Full text of "Recent advances in organic chemistry"

See other formats

i ,

RECENT ADVANCES IN
ORGANIC CHEMISTRY

BY THE SAME AUTHOR

STEREOCHEMISTRY

With 87 Illustrations. Crown %vo. los. 6d.

(TEXT-BOOKS OF PHYSICAL CHEMISTRY,
EDITED BY SIR WILLIAM RAMSAY, K.C.B., F.R.S.)

" The work is an important addition to the library of the physical chemist,
and as such forms a notable contribution to chemical literature." — The
Chemical Trade Journal.

" This book is practically a complete, and in many cases a detailed, account
of the subject of stereo or space chemistry since the foundations of this
extremely fruitful branch of chemical science were laid by Pasteur and
Wislicenus. It is not a historical summary, but a carefully thought out
treatise, and one which chemists who have to lecture or teach the subject will
find of the greatest use." — Nature.

" By the publication of this volume, Sir William Ramsay's series of
' Text-books of Physical Chemistry ' has received a valuable addition. Stewart's
' Stereochemistry ' is conceived in a thoroughly modern spirit, and can be
heartily recommended as a guide to the whole subject." — Physikalische
Zeitschrift.

"This volume maintains the high reputation of its predecessors. . . . The
countless references to original literature make the book a very valuable work
of reference." — Transactions of the Faraday Society.

"Though several treatises on the subject have already appeared, the author
has known how to handle it in an independent fashion of his own, and within
the limits of the volume has produced an extremely well-arranged work which
will serve as a guide to the present state of the science and as an inspiration to
further research."— PROF. W. OSTWALD in the Zeitschrift fur Physikalische
Chemie.

RECENT ADVANCES IN PHYSICAL
AND INORGANIC CHEMISTRY

With an Introduction by Sir WILLIAM RAMSAY, K.C.B., F.R.S.
&vo. 7-r. 6d. net.

"It is a volume which deserves the appreciation of all chemists." — The
Scotsman.

" If any one wishes to know how interesting the pursuit of science is,
especially chemistry, which after all is the science which deals with the most
ultimate of problems — the constitution of matter itself— we can recommend
nothing better than these two works." — Chemical Trade Journal.

LONGMANS, GREEN, AND CO.

LONDON, NEW YORK, BOMBAY, AND CALCUTTA

RECENT ADVANCES IN
ORGANIC CHEMISTRY

A. W. STEWART, D.Sc.

LECTURER ON ORGANIC CHEMISTRY IN THE QUEEN'S UNIVERSITY, BELFAST

WITH AN INTRODUCTION BY
J. NORMAN COLLIE, PH.D., LL.D., F.R.S.

PROFESSOR OF ORGANIC CHEMISTRY IN UNIVERSITY
COLLEGE, LONDON

SECOND EDITION

LONGMANS, GREEN, AND CO.

39 PATERNOSTER ROW, LONDON

NEW YORK, BOMBAY, AND CALCUTTA

1911

MY FATHER AND MOTHER

285780

PREFACE TO THE FIRST EDITION

IN the present volume the Author has aimed at giving a general
idea of the researches which have been carried out in Organic
Chemistry within the last ten years, but there has been no rigid
adherence to this period when it appeared desirable to include
earlier investigations. A considerable portion of the material
has not previously been collected in volume form ; and as far as
possible the most recent work in each branch of the subject has
been described.

Except in the case of the third * chapter, no attempt has
been made to enter into stereochemical questions. Eeaders
desiring information on this subject are referred to the Author's
book in Sir William Kamsay's series of "Text-Books of
Physical Chemistry."

In dealing with Organic Chemistry two courses are open ;
for we may consider the matter either historically or from the
synthetic point of view. In the present volume the second
method has been adhered to as far as possible ; and when the
synthesis of a substance is known, its constitution has been
deduced from the method of formation rather than from its
decomposition products. The latter, when important, are re-
served for consideration after the constitution has been demon-
strated. For the sake of clearness, each step in the syntheses
has been treated in a separate section, so that at any moment
the reader can see exactly how far he has advanced, and can
easily refer back to any stage which he may wish to read again.

As no one ever consults a book of this type when they wish
to know the boiling-point of a compound, it would have been
superfluous in the following pages to give more than the most
general account of the physical properties of the substances
mentioned. Full details on the subject are to be found in

* Chapter XI. in the present Edition.

viii PREFACE TO THE FIRST EDITION

Beilstein's " Handbuch der organischen Chemie," to which the
reader is referred for information on these points.

A bibliography of the subject has been appended to the
book, but it must be understood that it is not intended to be
complete; while the same may be said with regard to the
references at the foot of the pages. In both cases the aim has
been to furnish a guide to readers unacquainted with the
literature rather than to give a complete list of publications.

For the convenience of the reader, explanatory footnotes are
distinguished by asterisks, while references to the literature are
numbered. This, it is hoped, will obviate much unnecessary
reference to the foot of the pages.

Some chapters on the relations between chemical constitu-
tion and physical properties were originally projected ; but in
view of the approaching publication of a book on this subject
by Assistant-Professor Smiles in Sir William Eamsay's series,
it seemed unnecessary to go into the matter in this volume.

In conclusion, the Author desires to thank Professor Collie
for many suggestions made during the writing of the book ; and
especially for the Introduction which he has contributed. He is
much indebted also to Professor Inglis for improvements made
in the manuscript and for his kindness in reading the proofs of
the work.

A. W. S.

UNIVERSITY COLLEGE, LONDON,
September, 1908.

PREFACE TO THE SECOND EDITION

SINCE 1908, when this book was first published, further re-
searches have been carried out in several of the lines which
were then dealt with; and consequently in this new edition
considerable modifications have been made in order to bring
the volume up to date. .The original division of the alkaloids
into synthetic and unsynthesized types has ceased to be useful
owing to the number of these bodies which have recently been
synthetically prepared ; and on this account the two chapters
of the first edition have been fused into one. The Grignard
reaction, though novel in 1908, has now become so hackneyed

PREFACE TO THE SECOND EDITION ix

that there appears to be no advantage in devoting a chapter to
it ; this subject has therefore been omitted. The same fate has
befallen the chapter dealing with the chemical action of light.
At the time the last edition was published, this field of research
seemed to stand in need of a collected statement of the known
facts ; but at the present time the ramifications of the subject
are increasing to such an extent that it is almost impossible to
weave the disconnected data into a homogeneous whole ; and as
the space which the chapter occupied was required for other
questions, it seemed best to omit the subject from this volume.
Two new chapters, one on the quinoles, the other on the
triphenyl -methyl problem, have been introduced.

In the present volume the arrangement of the material
differs somewhat from that adopted in the first edition. Omit-
ting the first and last chapters, the book falls into two main
sections. The first of these, Chapters II.-VIL, includes the
syntheses a.nd constitution determination of important natural
bodies and their root-substances; while the latter half of the
book is devoted to compounds and problems which have some
bearing on our theoretical views.

In the preparation of this edition I have been much indebted
to Professor Inglis, Assistant-Professor Smiles, Messrs. Austin,
Clarke, Crymble, and Hilditch.

I should also like to tender my thanks to my numerous
reviewers both for their criticisms of the previous edition and
for their encouragement. When writing the first edition, I
tried to bear in mind that science is not a mere collection of
data, but is rather a rapidly changing series of hypotheses by
means of which we attempt to group the facts with which we
are acquainted ; and consequently I endeavoured (as one of my
reviewers put it, more clearly than I could do) " to illustrate
the principles upon which modern chemistry moves — not
stands — and to undermine the conservatism which exists in all
but strikingly original minds." The reception accorded to the
volume showed that this mode of regarding the subject is more
general than I had anticipated.

A. W. S.

THE STB DONALD CURRIE LABORATORIES,

THE QUEKN'S UNIVERSITY OF BELFAST,
October, 1910.

CONTENTS

PAOK

INTRODUCTION xiii

CHAPTER

I. MAIN CURRENTS IN ORGANIC CHEMISTRY DURING THE LAST

HALF CENTURY 1

II. THE POLY&ETIIYLENES 20

III. THE MONOCYCLIC TERPENES 38

1. Introductory 38

2. The Synthesis of Terpineol 39

3. The Decomposition Products of Terpineol 41

4. The Constitution of Dipentene 44

5. The Constitutions of Terpinolene and Terpinene .... 50

6. Terpin and Cineol 52

7. The Synthesis of Carvestrene 56

8. The Synthesis of Menthone 61

9. The Decompositions of Menthone 62

10. The Syntheses and Constitutions of Menthol and Mentherie 63

11. The Constitution of Pulegone 67

12. The Constitutions of the Phellandrenes 68

IV. THE DICYCLIC TERPENES 72

A. The Camphene Group 72

1. Syntheses of Camphoric Acid 72

2. The Synthesis of Camphor 75

3. Borneol, Camphene, and Camphane 76

4. The Decomposition Products of Camphor ..... 78

5. Camphoic and Apocamphoric Acids 81

B. Fenchone and its Derivatives 82

1. The Constitution of Fenchene 82

2. The Constitutions of Fenchone and Fenchyl Alcohol . 83

C. Pinene 85

1. The Constitution of Pinene 85

2. Pinonic and Pinic Acids 88

D. Bornylene and the Thujenes 89

CONTENTS xi

CHAPTER PAGE

V. THE OLEFINIC TERPENES 91

A. Introduction 91

B. Isoprene 91

C. Citronellal .. . i ..... 93

D. The Citral Group 96

1. General 96

2. Methyl-heptenone 96

3. Geranic Acid 98

4. Rhodinic Acid, Rhodinol, and Rhodinal 100

5. Citral 102

6. Geraniol, Nerol, and Linalool 106

VI. THE ALKALOIDS 110

A. General 110

B. Methods employed in the Determination of Alkaloid

Constitutions 113

C. The Pyridine Group 115

1. Coniine 115

2. Piperine 117

3. Trigonelline 119

D. The Pyrrolidine Group . . • . 120

1. Nicotine 120

2. Tropidine 124

3. Tropine, ^-Tropine, and Tropinone 127

4. Tropic Acid 130

5. Atropine 131

6. Ecgonine 132

7. Cocaine 133

E. The Quinoline Group 134

1. The Constitution of Cinchonine 134

2. The Constitution of Quinine 139

3. Cinchonidine and Conchinine 140

F. The Isoquinoline Group 140

1. The Constitution of Papaverine 140

2. The Synthesis of Papaverine 142

3. The Synthesis of Laudanosine 145

4. Opianic Acid 146

5. The Constitution of Cotarnine 147

6. The Synthesis of Cotarnine 151

7. The Synthesis of Hydrocotarnine 155

8. The Constitution of Narcotine 1 56

9. The Synthesis of Gnoscopine and Narcotine . . . 157

10. The Synthesis of Narceine 157

11. The Synthesis of Hydrastinine 158

12. The Constitution of Hydrastine 160

xii CONTENTS

CHAPTER PAGE

G. The Purine Group 161

1. The Synthesis of Uric Acid 161

2. The Synthesis of Theophylline 165

3. The Synthesis of Caffeine 166

4. The Synthesis of Theobromine 166

5. The Synthesis of Purine . . . .. 167

VII. THE POLYPEPTIDES 169

VIII THE POLYKETIDES AND THEIR DERIVATIVES 178

IX. THE QUINOLES 201

1. Introductory 201

2. Methods of Preparing Quinoles 205

3. The Properties of the Quinoles . 209

4. The Constitution of the Quinoles 213

5. Intramolecular Change in the Quinole Series 215

6. Conclusion 221

X. THE TRIPHENYLMETHYL QUESTION 222

1. Introductory 222

2. The Trivalent Carbon Hypothesis 225

3. The Hexaphenyl-ethane Hypothesis 228

4. Quinonoid Hypotheses 231

5. Tautomerism Hypothesis 237

6. Conclusion 240

XI. ASYMMETRIC SYNTHESES AND NEW METHODS OF PRODUCING

OPTICALLY ACTIVE COMPOUNDS 241

XII. SOME THEORIES OF ADDITION KEACTIONS 253

XIII. UNSATURATION 269

XIV. CONCLUSION 281

BIBLIOGRAPHY 290

NAME INDEX 297

SUBJECT INDEX . 301

INTRODUCTION

AT the present time it is not altogether easy to say on what
lines a text-book of Organic Chemistry should be written. To
state in the preface that the Author "hopes it will supply
a long-felt want" is not always an injudicious method of
announcing the Author's belief in the readers of text-books.
For if the " long-felt want " of the public is to have a restate-
ment of all the old facts once more, with nothing new, no critical
faculty shown, and an obvious lack of evidence that the book
can be used to broaden our outlook on other sciences as well as
chemistry, then no doubt the desire of the public for the time
being is satisfied.

It certainly is to be regretted, however, that so many books
on Organic Chemistry are published regardless of the fact that
Organic Chemistry is a growing science. If one wants to
know about a new piece of country, to obtain a large number
of photographs all taken from the same place is obviously
a foolish thing to do. Yet book after book on Organic
Chemistry is published, covering the same ground, with a fine
disregard of the fact that to the pioneers the outlook is con-
stantly changing. A book that has practically nothing new
in it except the description of a few more compounds is un-
necessary. Fortunately, however, there are some text-books
which are not mere narrations of facts, and which do point out,
not only what has been done, but what might be accomplished,
and which do make the reader think.

At no time, moreover, is a change wanted in the method
of writing text-books more than at present. Deluged as we
are with unnumbered facts that have often neither explanation
nor obvious connection with one another, Organic Chemistry
has become a vast rubbish heap of puzzling and bewildering
compounds. The sanguine chemist expresses a hope that some

xiv INTRODUCTION

day, perhaps, a few of these may be useful. All knowledge
ought to be useful, even that obtained by the manufacture of
the thousands of new substances which are annually produced
in chemical laboratories. But where is it to stop ? When one
looks at Beilstein's " Handbook" or Eichter's "Lexicon," or
takes up a current volume of any chemical journal, how many
of the compounds or the papers are of interest even to the
most enthusiastic chemist? The game of permutations and
combinations goes on, the chief object apparently being merely
to supplement the already existing myriads of laboratory-made
organic compounds.

How, out of all this undigested mass, is the writer of a text-
book to glean what is of interest or tell what ought to be taken
and what left? The result is that many text-books are not
much more than abridged chemical dictionaries. The only tax
on the reader's mind is to remember as many facts as possible.
The text-book is rare that stimulates its reader to ask, Why
is this so ? or, How does this connect with what has been read
elsewhere ?

Indeed, it is not inconceivable that a useful text-book might
be written on the constitutional formula of a single organic
compound ; for instance, alcohol. Its manufacture and physical
properties would have to be considered. This would necessitate
a knowledge of many typical organic compounds, and also of
many kinds of reactions. The evidence thus obtained could
then be summed up for the purpose of expressing all these facts
by the chemical formula. Here the theory of the constitution
of organic compounds would have to be dealt with, beginning
with the ideas in vogue at the beginning of last century:
Berzelius' Electro-chemical Hypothesis, of how the nature of
the elements present had the chief influence on the properties
of the compound; Dumas' Type-theory, and how he was the
first (about 1840) definitely to recognize the arrangement of
the atoms in the molecule : how this idea took about a quarter
of a century to get into the text-books ; how Frankland, in 1852,
started the idea of valency, from which sprang the modern
ideas of chemical structure and linking of atoms ; how Kekule
first definitely put forward the idea of the quadrivalence of
carbon ; how Crum Brown, in 1865, suggested the present form
of graphic formulae and pointed out that they were "not to

INTRODUCTION xv

indicate the physical but merely the chemical position of the
atoms." All these ideas have more or less centred round
alcohol and its derivatives; and any one who carefully had
followed the reasoning that led to these various mechanical
methods for representing by a chemical formula the molecular
structure of organic compounds would be in a position easily
to recognize that our present ideas must in future suffer change
just as they have done in the past.

Berzelius' ideas were those of a great mind ; but in his day
narrower theories were necessary for the more detailed develop-
ment of chemistry. Dumas' Type-theory, on the other hand,
was too narrow ; it was a very restricted system of classifica-
tion, and one that led to many false analogies. Up to the
present day, the Frankland-Kekule conceptions of valency and
graphic formulae have held their own, but there are signs that
these, too, will have to be modified ; physical as well as chemical
properties will have to be accounted for.

The present volume should be of great use to students of
organic chemistry. The subject-matter is put in an eminently
lucid form that enables the reader easily to follow all the
arguments, while at the same time his critical faculty is
stimulated. The book, moreover, is unlike so many modern
text-books in that it is not a mere compilation of facts ; several
novel theories on organic chemistry are dealt with, theories
that up to the present can hardly be said to have assumed
definite shape, but which rather point to the paths along which
the pioneers of the science are likely to go in the immediate
future.

J. NOKMAN COLLIE.

CHAPTEE I

MAIN CURRENTS IN ORGANIC CHEMISTRY DURING THE LAST
HALF CENTURY

SPEAKING exclusively of observational and experimental sciences,
it is obvious that progress can be accomplished only at the cost
of destroying or modifying current theories; for if a theory
suffices to explain facts discovered after its promulgation,
knowledge may be increased; but there is no true progress
unless our general outlook is altered. Thus in science we have
an alternation of two courses: in the first the aim is the
accumulation of facts and yet more facts ; while the second is
directed towards classifying these facts in the most convenient
manner. At irregular intervals some facts are discovered
which cannot be fitted into the accepted scheme of arrange-
ment, and in order to make room for them the scheme has
to be altered and recast into some new form.

In every science which is at all progressive there must
arise from time to time conflicts between the older generation
of workers and the leaders of the new ; for, to those who have
grown up along with it, a theory generally becomes invested
with a sort of sanctity which is quite out of keeping with
its true make-shift character. The longer a theory stands
the harder does it become to shake it, and the greater is the
tendency of the science to become stereotyped. There is
another side to the question. Without any disrespect, it may
be said that the majority of scientific investigators are not
possessed of strikingly original minds— we cannot all be
Pasteurs — and hence there is a very pronounced tendency to
take things as they are and work along the beaten track rather
than to push out in the wilderness and risk the chance of
losing the road altogether. Thus round every theory there
grows up a little band of workers, whose one aim seems to

3 • -R.ECEm": ADVANCES IN ORGANIC CHEMISTRY

be to accumulate evidence confirming their favourite hypo-
thesis ; and in this way the theory gains a considerable weight
of supporting data. On the other hand, the solitary worker
who happens to differ from the majority of his fellows has
to overcome a tremendous weight of unconscious prejudice
before he is able to obtain even the semblance of an impartial
hearing. In spite of these difficulties, however, progress is
made.

Chemistry has proved no exception to the general rule.
From the time of the phlogiston theory to the recent work
of Kamsay upon radium, the subject has been intermittently
developing, older theories have been reluctantly abandoned,
and a gradual change of standpoint can be traced, each advance
being forced upon the chemist by the impossibility of bringing
new facts into line with the older views.

When we examine the history of the origin and growth of
scientific theories it is curious to note how certain ideas seem
to pervade men's minds at a given period, though they may
remain unformulated for some years to come. Again and
again it has been found that two investigators have indepen-
dently pursued the same line of thought, and even accumulated
vast stores of facts with regard to the same subject, before any
suggestion has been put forward publicly. When we examine
these cases more closely we are often able to trace the evolu-
tion of the idea far further back than seemed possible at the
time ; the independent investigators themselves may have been
unaware of the existence of previous suggestions which bore
upon their views, but one can hardly avoid the view that
at given periods certain ideas are "in the air," having been
carried so far by previous workers that the new view forces
itself upon several minds simultaneously.

Such a crisis occurred in organic chemistry almost half
a century ago, when the foundations of our modern structural
theory were laid. Up to that time the theory of types had
served as a stop-gap, but it was too clumsy and inflexible to
respond to the ever-growing needs of a rapidly developing
science. Only those who have had occasion to refer frequently
to papers written previous to 1860, and who have been forced
to transliterate the older formulas into those employed at the
present day, can have any idea of the tremendous change

MAIN CURRENTS IN ORGANIC CHEMISTRY 3

brought about by the work of Frankland,1 Couper,2 and
Kekule.3 It is not necessary to enter into any discussion of
the part played by each of these in the growth of the modern
structure chemistry ; all three contributed an important share
to the theory upon which our work for the last fifty years has
been based. We may commence our present survey with the
period immediately preceding the publication of Kekule's first
paper on benzene.

Though the formulae of Couper and Kekule proved most
powerful weapons in the hands of those chemists who were
engaged in investigations of aliphatic substances, the equally
important class of aromatic bodies still remained in a state of
confusion equal to that which prevailed under the type theory.
It was the farsightedness of Kekule which brought this to an
end within less than a decade by a further advance along
structural lines. In 1865 he published a paper on this
subject ; 4 and a year later the whole problem was thoroughly
examined by him in a treatise5 which is probably as fine a
piece of reasoning as has yet been devoted to a chemical
question.

Kekule took as his first premise the fact that every aromatic
compound contains at least six carbon atoms ; secondly, when
a compound contains more than six carbon atoms it is often
possible to break it down into one containing six carbon atoms,
and further decomposition is resisted at this point, which appears
thus sharply to mark a definite stage in the process. From
these two points he was led to imagine that there was some-
thing in the arrangement of these six carbon atoms which
differentiated them from six carbon atoms grouped as in an
aliphatic substance. Another step completed the new theory.
Having advanced so far, Kekule had but to ask himself in what
way one could arrange six atoms so that they would not form
an open chain ; and it is now obvious to us that the simplest
reply is, in a ring. To us to-day, this seems such a self-evident
solution that we are apt to overlook the greatness of the
discoverer and to imagine that " any fool could have done it."

Frankland, Phil. Trans., 1852, 142, 417.
Couper, Phil Mag., 1858, IV. 16, 104.
Kekule, Annalen, 1858, 106, 129.
Kekule', Bull. soc. chim., 1868, 1, 98.
Kekule', Annakn, 1866, 137, 129.

4 RECENT ADVANCES IN ORGANIC CHEMISTRY

There is a certain element of truth in this, for it is apparent
that such a discovery might quite well have been the result of
a chance idea; in fact, Kekule himself uses language which
might give colour to this view, though probably he did not
intend to convey that impression. But Kekule was not content
with a mere statement of the problem's solution ; he was able
to forecast at once the line of research which would have to be
followed if the theory were to be put to the test of experience.

First, however, Kekule had to explain how the six carbon
atoms in the benzene ring could be linked together and united
with the six hydrogen atoms which are needed to make up the
complete benzene molecule. One of his early views was
speedily found to be untenable, as it presupposed two sets of
hydrogen atoms— three and three, so placed that a mono-
substituted benzene derivative might occur in two isomeric
forms. In its final form, the benzene ring was written practi-
cally as we write it now, with a double bond between every
alternate pair of carbon atoms and single bonds between the
other pairs.

At the time the benzene theory was developed, however, the
data which had been accumulated with regard to aromatic
compounds were not sufficiently numerous to establish definitely
its truth or error. But the new view imparted such a stimulus
to the investigators of that period that in a very short time it
was shown beyond a doubt that the Kekule theory was capable
of furnishing an interpretation of all the facts which had
previously been incapable of any clear arrangement.

No sooner had the benzene formula proved its value in this
way than a new problem was mooted. Given the benzene ring,
it is obvious, as Kekule himself pointed out in his paper already
referred to, that there must be a certain fixed number of isomers
for each substituted benzene derivative. For instance, if the
substituent introduced is always the same, there will be one
mono-substitution product, three di-substitution products, three
tri- substitution products, and so forth. The question then at
issue was the possibility of determining the constitution of any
given isomer ; or, in other words, if a poly-substituted benzene
derivative were produced in any reaction, what means must be
employed to discover the order in which the hydrogen atoms
and substituents lay round the ring ?

MAIN CURRENTS IN ORGANIC CHEMISTRY 5

The investigator who contributed most to the development
of this section of the subject was Kekule's pupil, Korner. If
we take a di-derivative of benzene and introduce into the ring
one additional substituent (thus forming a tri-derivative) it will
be found that the number of possible tri-derivatives depends
upon the constitution of the di-derivatives from which the
start was made. As can be seen from the figures below, an
ortho di-derivative will yield two tri-derivatives, a meta
di-derivative gives three tri-derivatives, while from a para-
compound only one tri-derivative is formed. The relations
of the tri-derivatives among themselves can be established by
an analogous method —

Korner, Griess, Ladenburg, Graebe, and Baeyer all aided to
establish the relations between the various substitution products
of benzene, and in a comparatively short space the filiation
between all the various classes of benzene derivatives had been
made clear.

It is a curious study to see how far one can trace in the
early controversies on the constitution of benzene the germs
of other theories which came later to their full development.
We may take one instance now. Ladenburg l was the first
to point out that while ortho-di-substituted benzene derivatives
occurred in one form only, the Kekule hexagonal formula
permitted two, which can be expressed by the formulae below.

1 Ladenburg, Ber., 1869, 2, 140.

6 RECENT ADVANCES IN ORGANIC CHEMISTRY

Iii the one case the carbon atoms carrying the substituents are
joined by a double, in the other case by a single, bond —

CH CH

HC CX HO CX

II I I II
HC CX HC CX

\/ v

Ladenburg adduced the supposed parallel case of crotonic
and vinyl-acetic acids, which differ in the position of the double
bond with reference to the carboxyl group —

CH3— CH:CH— COOH Crotonic acid.
CH2:CH— CH2— COOH Vinyl-acetic acid.

He alleged that if the shift of a linkage made no change in
benzene, it should be equally without effect in the case of these
two substances. But as they actually existed in isomeric forms,
the same was to be expected in benzene, if the Kekule theory
were correct.

Ladenburg was responded to by Kekule,1 and also by Victor
Meyer.2 The latter pointed out that while in benzene the
only difference between the two supposed isomeric forms was
produced by a mere change in the grouping of linkages, the case
of the two acids shown above was further complicated by the
fact that a hydrogen atom has also shifted its position from the
end of the chain to the carbon atom next the carboxyl group.
Thus the two cases are not really analogous at all.

Kekule attacked the Ladenburg view from a different stand-
point, and we cannot do better than quote his own expression
of the case —

" The atoms in the systems which we call molecules must be considered
to be continually in motion. This view lias often been expressed by phys-
icists and chemists, and I have already discussed it in my handbook.
Hitherto no explanation as to the form of this intramolecular motion has been
given, as far as I know. Chemistry must, at all events, bring forward the
claim that such a mechanical hypothesis accounts for the law of the linking
of atoms. A planetary motion seems, therefore, not inadmissible ; the move-
ment must be of such a kind that all the atoms forming the system retain

1 Kekule, Annalen, 1872, 162, 87.

2 V. Meyer, Annalen, 1870, 156, 265 ; 1871, 159,24.

MAIN CURRENTS IN ORGANIC CHEMISTRY ^

the same relative arrangement — in other words, that they return to a mean
position of equilibrium. If we now select from the numerous conceptions
which may be formed, that which gives the most complete account of the
chemical claims and which coincides most closely with the hypothesis of
modern physicists as to the mode of the motion of molecules, the following
assumption will appear the most probable.

" The simple atoms of the system approach each other in what may be
considered a straight line, and rebound like elastic bodies. What we call
valency would then have a mechanical meaning. Valency is the number of
contacts experienced by one atom on the part of other atoms in unit time.
In the same time that the monad atoms of a diatomic molecule strike each
other once, the dyad atoms of a diatomic molecule come twice into contact
with each other, the temperature being the same in both cases. In a mole-
cule made up of one dyad and two monads the number of contacts, under the
same conditions, in unit time is two for the dyad and one for each monad
atom.

" If two atoms of tetrad carbon are linked together by one combining
unit of each, they strike together once in unit time, or in the same time that
a monad hydrogen atom makes a complete vibration. During this time they
encounter three other atoms. Carbon atoms, which we call doubly linked,
come in contact twice in unit time, and further in the same period collide
twice with other atoms.

" If we now apply these views to benzene, the formula which I have pro-
posed appears as an expression of the following conception. Each carbon
atom strikes against two others in unit time, once against one and twice
against the other. In the same unit of time it comes once into contact
with hydrogen and completes one vibration.

" If we now represent benzene by the hexagon formula and fix our atten-
tion on one of the carbon atoms (that marked 1, for instance) —

(2) (3)
HG— CH

</ \
(1) HC CH (4)

HC=CH
(6) (5)

we can express the contacts it experiences during the first unit of time by—

(I.) 2, 6, h, 2,

where h stands for the hydrogen atom. In the second unit of time that
carbon atom which has just encountered No. 2 turns next to No. 6. Its
collisions during the second unit of time are —

(II.) 6, 2, A, 6.

While the contacts during the first unit of time are represented by the
hexagonal arrangement above, those during the second unit of time are
represented by —

8 RECENT ADVANCES IN ORGANIC CHEMISTRY

(2) (3)
HC=CH

(1) HC CH (4)

\ /
HC— CH

(6) (5)

The same carbon atom, therefore, is during the first unit of time doubly
linked to one of the adjoining carbon atoms, while during the second unit of
time it is in double linkage with the other carbon atom.

"The most simple mean of all the contacts of one carbon atom is found
from the sum of the contacts during the first two units of time, which then
repeat themselves periodically —

2, 6, h, 2, 6, 2, A, 6

and we see now that each carbon atom strikes against the two others with
which it is directly combined, an equal number of times ; in other words, it
bears the same relation to each of its neighbours. The ordinary formula for
benzene only represents the contacts made during the first unit of time, or
only one phase, and thus the view has sprung up that the di-derivatives with
the positions 1, 2 and 1, 6 must be different. If the above hypothesis or
any similar one be considered to be correct, it follows that this difference is
only apparent, not real."

Thus early in the history of modern structural chemistry did
the vibrational hypothesis make its appearance, and from that
time to the present day the view has slowly grown that the
intramolecular arrangement of atoms can best be represented
by a series of vibration phases rather than as a rigid assemblage
of particles.

The next stage in the evolution of this theory was taken in
view of quite different evidence. The Kekule benzene oscilla-
tion had been put forward to explain why two apparently
different structures had the same properties ; but in the question
of the acetoacetic ester constitution, which came to the front
soon after this, the crux of the problem lay in the fact that one
substance reacted as if it had either one or other of two totally
different structures.

In 1876 Butlerow x was led to suggest that in the cases of
certain bodies a kind of intramolecular vibration was continually
taking place, which explained why some substances reacted
now in one way and again in another according to the type of
reagent with which they were treated. Some years later, Laar 2

1 Butlerow, Annalen, 1870, 189, 76.

2 Laar, Ber., 1885, 18, 648; 1886, 19,730.

MAIN CURRENTS IN ORGANIC CHEMISTRY 9

collected a number of cases in which by shifting a hydrogen
atom from one atom to another in a chain it was possible to
give some idea of how the substances were able to act as if they
had two different constitutions. For instance, in the case of
acetoacetic ester it is found that with certain reagents it acts
as if it contained a hydroxyl group, while with other substances
it behaves as if it were a pure ketonic compound. This can
be expressed by saying that in the one instance it reacts as if it
had formula (I.), while in the other it appears to have the
structure (II.) —

(I.) CH3— C:CH— COOEt (II.) CH3-C— CH2— COOEt
OH O

This might be explained by supposing that what we call
acetoacetic ester is really a mixture of the two structure
isomers (I.) and (II.). Laar took a different view. According
to him, acetoacetic ester was a simple substance, but instead of
the hydrogen atom being attached either to the carbon or to the
oxygen atom it wandered or vibrated in space between them,
and was finally influenced in its choice of position by the action
of the reagent applied to the acetoacetic ester. We may
represent this by the following picture : —

CH3- C - CH— COOEt

Substances of this type Laar proposed to call " tautomeric "

uro, the same ; jue'/ooe, a part).

This idea of intramolecular vibration, however, soon
received an extension by the discovery of some cases in which
substances not only reacted as if they had two different struc-
tures, but could be actually isolated in the two structurally
distinct forms. This showed that in some cases at least the
Laar hypothesis was incorrect, or, rather, was too narrow a
statement of the case; for, instead of the wandering atom
remaining always like Mahomet's coffin midway between two
extremes, in these cases it was actually found at both ends of

io RECENT ADVANCES IN ORGANIC CHEMISTRY

the vibration range. A substitute for the Laar hypothesis was
brought forward by Jacobson.1 On his view, certain substances
may occur in either of two structurally different forms, and the
change of one form into the other takes place only under the
action of some reagent. Since any such change must be pro-
duced by a shifting of the bonds within the molecule, Jacob-
son proposed to describe the phenomenon by the name of
" desmotropy " (Setr/joc, bond ; rpo-rros, change).

Hantzsch and Herrmann 2 suggested that the whole subject
should be treated as one, so that if a substance could be isolated
in two stable forms it should be called desmotropic, while if it
could not be so isolated it should be termed tautomeric. We
need not go further into the question of desmotropy at present.
Enough has been said to show the growth of the idea of a labile
grouping of atoms within the molecule, which at the present
day has been carried as far as it can serve any purpose.

These views (which we may call " dynamic " in contradis-
tinction to the " static " conception of molecules as fixed group-
ings of atoms) did not come to a head in time to save Baeyer
from one of the greatest pieces of misdirected research which
the chemical world has seen in recent years. The work itself
is magnificent both from the practical and the theoretical stand-
point. But Baeyer carried out the whole of his investigations
upon one assumption, viz. that in the structural formula of
benzene there was a fixed, unalterable arrangement of valencies
which could be deduced from the results of oxidations, reduc-
tions, and other reactions. By this time many different modi-
fications of the original benzene hexagon had come into
existence, and Baeyer endeavoured to settle, by means of his
researches, which one of these actually represented the formula
of benzene. We cannot spare space to deal with the details of
his work, much of which has been of great service in directions
other than those in which it was originally aimed. Finally,
Baeyer himself was driven to conclude that there is no one
formula which will explain all the reactions of benzene. Collie 3
has shown how all the proposed benzene formulae may be
harmonized and expressed by a simple vibrational system in

1 Jacobson, Per., 1887, 20, 1732; 1888, 21, 2628.
- Hantzsch and Herrmann, Ber., 1887, 20, 2803.
3 Collie, Trans. Chem. 8oc., 1897, 71, 1013.

MAIN CURRENTS IN ORGANIC CHEMISTRY 11

three dimensions, and it seems unlikely that this will be
improved upon to any great extent.1

From his examination of the benzene question, Baeyer was
led to take up the subject of the terpene constitutions, which
at that time was in its infancy. These substances are in most
cases derivatives of reduced benzene rings, so that Baeyer
passed from the one subject to the other quite naturally. To
his work in that line we owe much of our present knowledge of
terpene chemistry ; but we are even more indebted to Wallach,
who began work in this subject about the same time as Baeyer.
We cannot give even the briefest summary of Wallach's work
in the space at our disposal here, but must content ourselves
with referring the reader to special treatises on the terpenes
and ethereal oils.

This brings us to the camphor controversy, which for a
decade raged through a corner of the chemical world. Every
organic chemist of note seems to have considered himself in
duty bound to propose some formula for camphor or a camphor
derivative, and the confusion resulting from this prodigality was
only banished by the synthesis of camphoric acid and camphor,
which we shall describe in the chapter upon the dicyclic terpenes.

The reader will now have some idea of the extraordinary
fertility of the theory of aromatic compounds put forward by
Kekule. We must next turn to another question in which
Kekule, if not the actual originator, was at least one of a long
chain of investigators whose work has had a tremendous
influence upon our ideas of intramolecular arrangement.

When one looks back upon the work of scientific discovery,
one is struck most, not by the fact that certain things have
been discovered, but by the very slightness of the barrier which
so often stood between the success and failure of a certain line of
research at a given period. Again and again subjects have been
approached and their problems virtually solved, yet for want of
just one connecting link, or even the addition of a few words to
a statement which in itself contains the key to the problem, the
question may go unanswered for years. No better example
of this is to be found than that furnished by the evolution of
stereochemical theory.

1 A discussion of the various space formulae for bcnzeue will be found in the
author's " Stereochemistry."

12 RECENT ADVANCES IN ORGANIC CHEMISTRY

In 1860 Pasteur1 carried out an investigation of the
tartaric acids, in which he was able to show that crystals of
dextro- and Isevo-tartaric acid are related to one another as an
ohject is related to its image in a mirror. At this time the
structure theory was in its very infancy, and Pasteur does not
seem to have thought of applying it to the case with which he
was dealing. He contented himself with putting forward as
a possible explanation the view that the atoms in the tartaric
acid molecule were arranged in right- or left-handed spirals, or
were grouped at the corners of a tetrahedron. This was the
germ of the whole of modern stereochemistry, but, for want of
a slight addition to these expressions, it remained for later inves-
tigators to reap the credit of establishing the correctness of
this view. In 1869 Paterno2 proposed to explain certain cases
of isomerism by means of tetrahedral models. Kekule,3 two
years previously, had described a tetrahedral model, but it
seems doubtful whether he really intended it to convey an idea
of the distribution of valencies in four directions in space. No
notice was taken of either of these suggestions by the chemical
world in general, and it appears to have been Wislicenus 4 to
whom we owe the next definite pronouncement on the subject.
After proving that the structures of the isomeric lactic acids
were identical, he added, " The facts force us to explain the
difference between isomeric molecules of the same structure by
a different arrangement of atoms in space."

The ultimate solution was published simultaneously by
Le Bel 5 and van't Hoff,6 who pointed out that all organic
substances showing optical activity contained at least one
asymmetric carbon atom, i.e. an atom whose four valencies are
attached to four dissimilar groups. A slight extension of the
theory sufficed to explain the occurrence of isomeric substances
containing a double bond ; and Baeyer 7 applied it also to the

1 Pasteur, " Recherches sur la dissymmetric inoleculaire des produits
organiques naturels."

2 Paterno, " Giorn. di Scienze Naturali ed Econ." V. Palermo, 1869.

3 Kekule', Zeitsch.f. Chem., 1867, N.F., 3, 217.

4 Wislicenus, Annalen, 1873, 167, 343.

5 Le Bel, Bull soc. chim., 1874, II, 22, 377.

6 Van't Hoff, " Voorstell tot uitbreiding der structuur formulas in de ruinate."
Utrecht, 1874.

7 Baeyer, Ber., 1885, 18, 2277.

MAIN CURRENTS IN ORGANIC CHEMISTRY 13

case of cyclic substances. But the theory has been by no
means limited to carbon compounds alone. In 1890 Hantzsch
and Werner1 had recourse to stereochemical ideas to explain
cases of isomerism among the oximes; in 1893 Werner was
able to bring some semblance of order into the class of metal-
ammonia derivatives; while in 1894 Hantzsch2 put forward a
theory of the diazo-group. On the side of the question dealing
with optical rotatory power, the work of Le Bel 3 on nitrogen,
Smiles4 on sulphur, Pope and his students5 on selenium and
tin, and Kipping6 on silicon, have shown that asymmetric
atoms of these elements may also give rise to activity.

So much for the statical side; but there is another point
of view from which we may regard the relations between the
positions of atoms in space. In this new field Victor Meyer and
Bischoff have contributed by far the greatest additions to our
knowledge. They have shown that reactions may be hindered, or
even completely impeded, by certain groupings of atoms in given
positions. For example, if in benzoic acid we substitute methyl
groups for the two hydrogen atoms in the ortho-positions to the
carboxyl group, the acid becomes at once much more difficult
to esterify. This phenomenon is termed " steric hindrance."

We have not space to enter into any question of stereo-
chemistry in detail at present ; but in this connection we must
mention one of the greatest pieces of research which have been
carried out in the past twenty years. When Emil Fischer and
his students first began methodically to examine the sugars, the
investigation of that class of bodies was regarded as one of
the most hopeless problems which an organic chemist could
set himself. The substances were often uncrystallizable, and
differed so little among themselves that it seemed hopeless
to try to separate one isomer from a mixture. Further, the
enormous complication of their isomerism, due to the numerous
asymmetric carbon atoms they contain, seemed to make the
attack upon this branch of stereochemistry one of the least

Hantzsoh and Werner, Ber., 1890, 23, 11.
Hantzsch, Ber., 1894, 27, 1702.
Le Bel, C. B., 1891, 112, 724; 1901, 129, 548.
Smiles, Trans. Chem. Soc., 1900, 77, 1174.

Pope and Peachey, Proc. Chem. Soc., 1900, 16, 42, 11G; Pope and Neville,
Trans. Chem. Soc., 1902, 81, 1552.

6 Kipping, Tram. Chem, Soc., 1907, 91, 209, 717; 1908, 93, 457 ; 1909, 95, 69.

14 RECENT ADVANCES IN ORGANIC CHEMISTRY

promising. There seems no doubt that with ordinary means
at his disposal Fischer could never have accomplished the
work ; but his genius had stood him in good stead in one of his
earlier researches. In 1877 he discovered the compound
phenylhydrazine,1 and by means of this substance he was able
to convert the imperfectly crystallizable sugars into crystalline
hydrazones, from which he was able to regain the sugar after
he had thus separated it from its isomers. In a few years
Fischer2 completed this vast research, in the course of which
he established the configurations of all the pentoses and hexoses
by experimental means and by reasoning which is unlikely to
be surpassed for simplicity and directness.

After leaving the sugars, Fischer devoted his attention to
the purine group, in which he carried out a series of brilliant
syntheses ; and when this subject in its turn was exhausted he
attacked the problem of the decomposition products of the
peptones, with results which are described in a later chapter of
the present volume.

There are one or two other problems which have been dealt
with in the last thirty years, but they are somewhat dis-
connected with each other and with the parts of organic
chemistry which have just been described.

In the first place, there is the pyridine question. After the
constitution of benzene had been established, it was inevitable
that the same view would sooner or later be applied to pyridine,
and in 1869 Korner3 proposed to represent that substance by
a benzene ring in which one of the — CH= groups was
replaced by a nitrogen atom. This theory was supported by
some researches of Kekule,4 and is to-day accepted as correct.
Now, the importance of pyridine and its simple derivatives does
not lie in themselves, but rather in the fact that the pyridine
ring appears to form the basis of all the natural alkaloids, just
as the benzene ring is the foundation of the aromatic series.
We need not enter into the alkaloid question here, as a
chapter in this volume is devoted to it.

1 Fischer, Annalen, 1877, 190, 81.

2 Fischer, Ber., 1894, 27, 3189.

3 The first publication of this idea seems to be due to Dewar, Zzit.f. Chem.,
1871, 117.

4 Kekule', Ber., 1890, 23, 564; £ee also Bichter-Anschutz, Lahrbuch d.
Organ. Chemie, 1905, II. 711, 712.

MAIN CURRENTS IN ORGANIC CHEMISTRY 15

Since the time of Kekule, organic chemistry has been for
the most part a synthetic science. At the present day con-
siderably over a hundred thousand organic compounds are
known, and one need not have the least hesitation in saying
that if seventy per cent, of them had never been synthesized
we should not feel the lack of them to any appreciable extent.
The reason for this enormous flood of synthetic material is to
be found in the German University system ; for since, under the
German regulations, the degree in chemistry is granted only on
the results of original research, it follows that every Ph.D.
represents so many new compounds — at least, as a general
rule. But these do not include all the forces leading to the
steady pursuit of the synthetic branch. The great German dye
industry employs in itself hundreds of chemists, and from them
also flows a steady stream of new compounds. The same may
be said of the explosive manufacturers and the firms which
produce synthetic drugs.

Before closing this chapter we may cast a glance at the
physical methods which have sprung up in organic chemistry
during the last half-century. The relation between chemical
constitution and optical rotatory power dates, of course, from
the time of van't Hoff and Le Bel's papers on the asymmetric
carbon atom; and Guye1 has propounded a theory which,
though failing in detail, seems not inapplicable to the general
connection between rotation and constitution. Eefractive index
appears to be a property which is closely connected with the
mode of linkage of the atoms in organic compounds, and much
work has been done in this line by Briihl and others. The
electrical conductivities of acids depend very greatly upon the
constitution of the radical to which the carboxyl group is
attached. Magnetic rotation, i.e. the optical rotatory power
which nearly all substances acquire when placed in strong
magnetic fields, has been studied in great detail by the late
Sir W. H. Perkin, who showed that by its aid the constitution
of many substances, especially desmotropic bodies, could be
determined. Absorption spectra, both optical and electric, have
been used in the determination of doubtful structures ; the first
have been employed by Hartley, the second by Drude.

We have now completed our survey of modern organic

1 Guye, C. It,, 1890, 110, 714.

1 6 RECENT ADVANCES IN ORGANIC CHEMISTRY

chemistry. It has, of course, been impossible to deal with
many pieces of work which are quite as important as some of
those we have actually mentioned, but, on the whole, it is
believed that a more or less accurate picture has been given of
the evolution of the subject along various lines. If we look
only to the principles which lie at the back of the science and
which, though they may remain unformulated, still sway our
views by some sub-conscious action, we shall find that the
history of the last five and twenty years has been one of a
gradual passing from a static to a dynamic view of the molecule.
In the early days, the ideas of chemists centred round more or
less rigid structures which they regarded as approximately
"set." The tetrahedron of van't Hoff had the effect of
strengthening rather than weakening this tendency ; and it is
much to be regretted that the van't Hoff view, rather than
that of Le Bel, found favour in the chemical world at large.
At this period the state of mind1 of the average organic
chemist seems to have been somewhat similar to that of the
student who, when asked to explain the atomic theory, said,
" Atoms are square blocks of wood invented by Dr. Dalton."

This view of the subject was shaken by the publication of
Werner's views on affinity and valency,2 in which the idea
of directed valencies was shown conclusively to be a quite
unnecessary assumption ; and at the present day the idea of a
certain amount of intramolecular " play " is not regarded as
absolute anathema by the more advanced school.

The last twenty years of organic chemistry, however, have
been rather barren in many directions. The only really out-
standing conception which has been evolved and developed
into several branches has been the idea of the arrangement of
atoms in space. Apart from this, the theoretical side of the
subject has not given rise to anything more than a series of
very minor theories, none of which (with the exception of
Thiele's partial valency hypothesis) seems likely to develop
in any wide manner. The reason for this is most probably
to be sought in the recent and sudden rise of physical
chemistry, which has drawn away from the organic field many

1 If this statement appears exaggerated, the reader is recommended to
consult the Annalen, 1901, 316, 71, where he will find ample evidence of the
persistence of these views even quite recently in certain circles.

2 Werner, " Beitrage zur Theorie der Affinitat und Valenz." 1891.

MAIN CURRENTS IN ORGANIC CHEMISTRY 17

chemists who would doubtless have carried the older branch
much further forward than has been possible without their
assistance.

The progress of human thought has been likened by some
author to the journey of a passenger on a funicular railway.
We leave the starting-point in the valley and pass upward
through tunnels and cuttings for a time, until at last we emerge
again into daylight to find that though we are still within a
stone's throw of our point of departure we have moved in a
spiral, and now look down from a new point of view at the
station which we have left. In the same way, science appears
to progress in cycles; and, after a more or less prolonged
period, we find the old views reappearing and the old con-
flicts beginning once more, though at each new encounter
the point of view is slightly shifted as new or more refined
experimental methods replace or supplement the older ones.

This periodicity in theory has seldom been displayed more
clearly than in the matter of structural formulae and their
meaning. If we examine the views of the two pioneers Couper
and Kekule, we find that, though agreed as to the method of
writing down formulae, they were by no means at one as
regards what the formulas expressed when once put upon
paper. Kekule's view * was based upon quite incontrovertible
reasoning : —

" Rational formulse are decomposition formulse, and in the present state
of chemical science can be nothing more. These formulse give us pictures
of the chemical nature of substances ; because the manner of writing them
indicates the atomic groups which remain unattacked in certain reactions.
. . . Every formula which expresses definite metamorphoses of a compound
is rational; that one of the different rational formulae is the most rational,
which expresses the greatest number of these metamorphoses."

Couper,2 on the other hand, put the case as follows : —

" Gerhardt ... is led to think it necessary to restrict chemical science to
the arrangement of bodies according to their decompositions, and to deny the
possibility of our comprehending their molecular constitution. Can such a
view tend to the advancement of science ? Would it not be only rational, iu
accepting this veto, to renounce chemical research altogether ? "

Thus, on the one side, we have Kekule maintaining that

1 Kekule, AnndUm, 1858, 106, 149.

2 Couper, Phil. Mag., 1858, IV. 16, 107.

18 RECENT ADVANCES IN ORGANIC CHEMISTRY

graphic formulae are simply shorthand symbols by means of
which we can easily and compactly express the results of our
chemical experiments ; while, on the other side, Couper claims
that these ciphers give us the key to the actual mode of linkage
of the atoms within the molecule.

These two theories still persist side by side in the present-
day chemical world, and a considerable waste of energy occurs
when two upholders of different views try to interpret the same
set of results. The conflict between the two schools has become
especially acute within recent years owing to the strides which
have been made in the correlation of chemical constitution and
physical properties of compounds ; for a substance may react
as though it possessed a given constitution, whereas physical
measurements would lead us to quite different conclusions with
regard to the atomic linkages of the molecule in question. In
the case of a chemical problem, it seems evident that chemical
evidence should carry most weight, and the confusion of ideas
has been chiefly due to workers on the physical side employing
chemical formulae to denote something which is not chemical.
It seems a peculiar process of reasoning which allows a person
to base his hypotheses upon the reaction formulae of chemists,
and then, after taking into account some physical property of a
substance, to return to chemistry with a proof of the incorrect-
ness of the original reaction formulae upon which his argumen-
tation is based.

When we look at the present condition of organic chemistry
it is rather discouraging. Everything seems to be cut and
dried as far as the theory of the subject is concerned, and on
the practical side the main tendency seems to be merely to
confirm current theories. It is self-evident that we can make
no progress by confining ourselves to the confirmation of the
views which pass muster at present. If any advance is to be
made, it must be begun by investigating those phenomena which
do not agree with the standard theory ; and, as an instance, we
might point to the case of the ionic hypothesis and the ordinary
organic reaction. There is, however, another direction in which
we may look for a line of research. It is customary in text-
books to assume that " water adds on in such and such a way,"
or, " the bromine atom attaches itself in the usual way to the
tertiary carbon atom " ; many such expressions can be found,

MAIN CURRENTS IN ORGANIC CHEMISTRY 19

but it never seems to occur to the ordinary person that to state
a fact is not to offer an explanation, and while we are all suffi-
ciently glib in describing how a reaction takes place, very few
of us seem to give a thought to the problem of why the reaction
takes that particular course rather than another. We have
thus accumulated an immense mass of data concerning the
results of reactions, but very little indeed with regard to their
causes. It seems obvious that if organic chemistry is to get a
new lease of life, some attention must be paid to such questions
as these. Of course the investigator who takes up such pro-
blems will have to invent a new set of methods ; but the aim
in view would be worth a little trouble.

At the present day it appears to be the fashion to suppose
that certain views are so firmly established that no research
into their foundations is worth the labour expended on it, and
consequently investigators devote much time and energy to the
examination of highly complex substances while simpler com-
pounds are supposed to be " worked out." In the same way it
was supposed for many years that the composition of the
atmosphere was well known, until the work of Eamsay and
Eayleigh showed how little we knew of even this common
mixture. With this object lesson before them, it is to be
hoped that more organic chemists will find time to investigate
some of the problems which are passed over by the mass of
workers who seem to place a label in the same category as an
explanation.

CHAPTER II

THE POLYMETHYLENES

IN the succeeding chapters we shall deal with the mono- and
di- cyclic systems which are found among the terpenes and
camphors; but before entering upon a discussion of these it
appears desirable to give some account of the root-substances
from which all of them are derived. The present chapter,
therefore, will be devoted to the polymethylenes.

Compounds of this class have the general formula (CHa),,,
where "n" is any integer from three to nine; and they are
therefore isomeric with the open-chain olefinic compounds of
the general formula CnH2n. The two classes differ widely
from each other, both in chemical and in physical properties ;
this point will be discussed in detail later in the chapter.

Two nomenclatures are at present in vogue for derivatives
of this class of compounds. In the first, the given substance is
distinguished as a tri-, tetra-, penta- hexa-, hepta-, octo-, or nono-
methylene according as its ring is made up of three, four, five,
six, seven, eight, *or nine methylene groups. If a carbonyl
group occurs in the ring, its presence is indicated by the prefix
" keto-," while for a carboxyl group the suffix " -carboxylic acid "
is added to the name of the polymethylene. The second system
of nomenclature is a more general one. The designation of
any polymethylene is found on this second system by taking
the name of the corresponding paraffin and prefixing " cyclo- "
to it. When a double bond occurs in the compound the
termination " ane " is changed to " ene " ; and for two double
bonds to " di-ene." If a ketonic group occurs in the molecule
it is distinguished by changing the termination "ane" to
" anone." As can be seen, both systems are somewhat clumsy,
and hence it is desirable at times for the sake of clearness
to use the one which most simply expresses the compound

THE POLYMETHYLENES 21

in question. The following examples will help to make the
matter clearer :—

CH2

CH2 CH2— CH CH CH

/ \ I II II II

CH2 - CH2 CH2— CH CH - CH

Trimethylene Cyclobutene. Cyclopentadiene.

Cyclopropane.

CH2— CH.COOH CH2 CH2

OH — Oil

Tetramethylene-carboxylic acid. Ketopentaraethylene

Cyclobutane-carboxylic acid. Cyclopentanone.

There are at present ten principal methods by which we can
obtain saturated cyclic carbon compounds ; of these, only four
yield simple polymethylenes ; two others produce homologues of
the parent substances ; and the remaining methods lead to the
formation of acids with a polymethylene nucleus. We may
examine all these reactions, and for the sake of convenience in
future reference it may be well to number them consecutively.

(1) The simplest method of obtaining a polymethylene com-
pound is to act upon the corresponding open-chain dihalogen
derivative with zinc dust or sodium. This is merely a modifi-
cation of the ordinary Fittig-Wiirtz reaction—

)Ho

CH2Br CH2

a= 21sraBr4.H2C
CH2Br CH2

(2) When the calcium salt of a mono-basic acid is distilled
it yields calcium carbonate and a ketone. The same reaction
was employed by Wislicenus and Hentzschel l in the case of a
dibasic acid; and the resulting compound was found to be a
cyclic ketone —

1 Wislicenus and Hentzschel, Annalen, 1893, 275, 312.

22 RECENT ADVANCES IN ORGANIC CHEMISTRY
CH2.CH2.CO.O CH2.

Ca = CaCOo +

CH2.CH2.CO.O CH2.C1

From the ketone the corresponding secondary alcohol was pre-
pared by reduction with sodium in ethereal solution : and from
the alcohol, by the action of hydriodic acid at 0° C., the iodide
was formed. This, on reduction with zinc and hydrochloric
acid, gave the corresponding hydrocarbon —

. OH2 \j xi 2 . Cxi 2

CH . OH >

CH2 . C

CH2

CH.I

CH2 . CH2

CH2 . GH2

CH,

CH2.C

(3) If the unsaturated cyclic hydrocarbon corresponding to
the desired polymethylene is known, the saturated compound
may be obtained from it by passing its vapour, mixed with a
stream of hydrogen, over finely divided nickel. This method
was devised by Sabatier and Senderens.1 The nickel is heated
while the gas is passed over it, the temperature being regulated
with care, as the action is apt to be carried too far and to lead to
the opening of the ring by further reduction of the polymethylene.

(4) In those cases in which it is possible to obtain the
amine derived from the required polymethylene, it can be con-
verted into the parent substance by Kishner's method.2 In
the first place, the amine is converted into the hydrazine by
bromination and subsequent treatment with silver oxide ; the
hydrazine is then oxidized with alkaline potassium ferricyanide.
Nitrogen is finally evolved, and the required hydrocarbon
remains behind. For the sake of simplifying the formulae we
may take the theoretical case of the production of tetramethy-
lene from amidotetramethylene ; the steps in the reaction are
indicated below —

1 Sabatier ani Senderens, C. R., 1901, 132, 210.

2 Kishner, J. pr. Ch., 1895, II., 52, 424.

THE POLYMETHYLENES

J? w?

£3 O

W W
o— o

— O

S-o

0—0

M-^ •— —

o—o

8-8

— - "~~"
o—o

hri &3
0—0

24 y RECENT ADVANCES IN ORGANIC CHEMISTRY

(5) The foregoing methods lead direct to the simple poly-
methylenes, but homologous compounds may be prepared in
other ways. For instance, the reaction of pinacone formation
may be modified in such a way as to give us cyclic alcohols.
In the case of ordinary pinacone syntheses two molecules of
a ketone unite during the reduction process. If for the two
ketone molecules we substitute a single molecule of a diketone
the reaction-product will in this case also be a pinacone ; and
in some cases two pinacones are formed, one an open-chain
compound, the other a polymethylene derivative. For example,
Kipping and Perkin,1 by the reduction of diacetyl-pentane,
obtained a mixture of dihydroxy-nonane (I,) and dimethyl-
dihydroxyheptamethylene (II.).

,CH2 . CH2 . CH(OH) . CH3
CH2 (I.)

/CH2 . CH2 . CO . CH3 / \CH2 . CH2 . CH(OH) . CH3
CH

CH2 . CO . CH3^ /CH2 . CH2 . C(OH) . CH3

CH2

\CH2.CH2.C(OH).CH3

(II.)

The hydrocarbon may then be prepared from the pinacone
in the usual way.

(6) The Grignard reaction has been applied to the pro-
duction of polymethylene homologues by Zelinsky and Moser,2
who prepared methylpentamethylene from w-aceto-butyl iodide
by the action of magnesium. The reaction takes place in the
following steps : —

CHg.CO

H2C
H,n

CH2

CHg.CO

H2C
HP

Mgl
CH2
ill

CHg.C.O.Mgl
H2C vx'Jtl2
Hn PTT,,

2
CH3.

H*C
TT_P

C.OH
CH2

PIT-

CH3
KsC

.C.H

/\
CH2

PR.

1 Kipping and Perkin, Trans. CJiem. Soc., 1890, 57, 241.

2 Zelinsky and Moser, Ber., 1902, 35, 2684.

THE POLYMETHYLENES 25

(7) Buchner and Curtius l were the first to point out that
the aliphatic diazo- compounds had the faculty of coupling with
unsaturated substances to yield pyrazole or pyrazoline deriva-
tives. These latter bodies, on distillation, break down into
nitrogen and trimethylene compounds. For example, in the
case of diazomethane and fumaric ester the reaction takes the
following course : —

CH2 CH.COOEt CH2— CH.COOEt

\ + II =1 I

N = N CH.COOEt N CH.COOEt

v \

. COOEt
COOEt

Nje,

(8) The remaining methods with which we have to deal
depend upon such reactions as the acetoacetic or malonic ester
condensations. It is obvious that just as we obtained a cyclic
ketone by substituting the calcium salt of a dibasic for that of
a monobasic acid, we could obtain an intramolecular condensa-
tion by substituting for acetic ester the ester of a dibasic acid.
For example, if we used adipic ester, and proceeded in the
same way as in the ordinary acetoacetic ester synthesis, we
should obtain a keto-pentamethylene carboxylic ester —

CH2.CO.OEt
| - Eton

i i — > CH2.CO.OEt

OH.. <!o

CH2.CH2.CH.CO.OEt CH2 . CH2 . CH . CO . OEt

| -Eton i I

|H I — > CH2 CO

CH2— COJ OEt I

(9) Again, if we condense oxalic ester with a dicarboxylic
ester by means of sodium ethylate we can obtain a diketo-
polymethylene dicarboxylic ester —

1 Buchner and Curtius, B&r., 1885, 18, 237.

26 RECENT ADVANCES IN ORGANIC CHEMISTRY
CO j OEt H^-OH . COOEt CO CH . COOEt

\ -2EtOH

CH2

0 i OEt H4-CH . COOEt CO CH . COOEt

(10) The last series of methods1 which we need describe
depends upon the interaction of alkyl halogen compounds with
the sodium derivatives of malonic, acetoacetic, or ethylene
tetracarboxylic ester. Some examples will serve to make the
matter clear —

CH2.Br COOEt CH2 COOEt

C C + 2NaBr

COOEt CH2 COOEt

CH2.Br

CH2.Br Na . CH . COOEt CH2 CH. COOEt

| | =||+ NaBr

CH2.Br CO.CH3 CH2.Br CO.CH3

This bromine compound may now react with more sodium
ethylate in either of two ways, yielding in the one case a
trimethylene derivative, and in the other an internal ether —

COOEt COOEt

I I

CH2 C— CO . CH3 > CH2— C— CO . CH3

CH2.BrNa CH2

Ketonic form.

CH2 C . COOEt CH2 C . COOEt

I II > I II

CH2Br NaO . C . CH3 CH2— 0— C . CH3

Enolic form.

The reaction between an alkylene dibromide and the disodium
derivative of ethane tetracarboxylic ester takes place as
follows :—

CH2.Br Na.C(COOEt)2 CH2— C(COOEt)2

I 4- =| +2NaBr

CH2 . Br £Ta . C(COOEt)2 CH2— C(COOEt)2

1 Perkin, Ber., 1884, 17, 54 ; Perkin and Freer, Tram. Chem. Soc,, 1887, 51,
833; Baeyer and Perkin, Ber., 1884, 17, 448 ; Perkin, Trans. Chem. Soc., 1888,
53,1.

H = CH

THE POLYMETHYLENES 27

This method may be modified by substituting for the ethane
tetracar boxy lie ester an alkylene dimalonic ester and using
iodine instead of the alkyl halide —

Na . C(COOEt)2 C(COOEt)2

Ia+ C

Na . C(COOEt)2 C(COOEt)2

We must now deal with the individual members of the
polymethylene series.

The simplest member of the group, trimethylene, was dis-
covered by Freund,1 who prepared it by the action of sodium
upon trimethylene bromide —

CH2.Br CH2

-f 2NaBr

CH2.

H2 + Naa = CH2

Br CH2

It is a gas at ordinary temperatures, melts at —126° and
boils at -35° approximately under a pressure of 749 mm.2
Trimethylene is isomeric with propylene, from which it can be
distinguished by means of halogens or halogen acids. In the
case of the polymethylene, chlorine breaks the ring and pro-
duces trimethylene chloride ; while propylene takes up chlorine
to form propylene dichloride—

CIL

I- C12=CH2
CH2 CH2C1

The next member of the series, tetramethylene, has been
known only within recent years, though for twenty years
attempts to prepare it had been made by various workers, but
without success. It was at last produced by Wills tatter
and Bruce 3 in the following way. Tetramethylene carboxylic
acid (I.) was first prepared, for it should be noted that though

1 Freund, Monatsh., 1882, 3, 625 ; J. pr. Ch., 1885, II. 26, 367.

2 Ladenburg and Kriigel, Ber., 1899, 32, 1821.

3 Willstatter and Bruce, Ber., 1907, 40, 3979.

28 RECENT ADVANCES IN ORGANIC CHEMISTRY

the parent substance was unknown, many tetramethylene
derivatives had been prepared by the general methods given
above. This acid was converted into the amide (II.), and this,
by the usual reaction with bromine and soda, gave the amine
(III.). From this, by methylation, tetramethylene-trimethyl-
ammonium hydroxide (IV.) was obtained, which, on distillation,
broke down into various compounds, the only one which concerns
us being the cyclobutene (V.). When this body is reduced by
the Sabatier and Senderens method, passing it with a stream of
hydrogen over nickel powder at a temperature not exceeding
100°, it is converted into tetramethylene (VI.).

CH2 . CH . COOH CH2 . CH . CO . NH2 CH2 . CH . NH2

0x12 • 0x12 Ori2 . Oil2 0x12 . OlJ2

(I.) (II.) (III.)

CH2.CH.N(CH3)3OH CH2.CH CH2.CH2

CH2.CxI
(IV.) (V.) (VI.)

Tetramethylene is a gas at ordinary temperatures, but con-
denses to a liquid with a boiling-point of 11°-12° C. It does
not solidify even at - 80° C.

Pentamethylene was first produced by Wislicenus and
Hentzschel by the second general method given above. It is
a light liquid, boiling at 50° C., and remaining unsolidified at
-80° C. It is found to occur naturally in Caucasian1 and
also in American 2 petroleum.

Hexamethylene was first synthesized by Baeyer3 by the
reduction of 1, 4-diketo-hexamethylene. Perkin and Haworth 4
produced it by the action of sodium upon a boiling alcoholic
solution of di-bromo-hexamethylene, and it is also formed by
the reduction of benzene. It is a pleasant-smelling liquid,
boiling at 81° C., and solidifying to a solid of melting-
point 6° C.

1 Markownikoff, Ber., 1897, 30, 974.

2 Young, Trans. Chem. Soc., 1898, 73, 906.

3 Baeyer, Annalen, 1894, 278, 111.

* Perkin and Haworth, Ber., 1894, 27, 216.

THE POLYMETHYLENES 29

Heptamethylene, or suberane, was obtained by Markowni-
koff l from suberic acid (I.) by means of the second general
method. The distillation of calcium suberate gave suberone
(II.), which was reduced to suberyl alcohol (IIL), from which
suberyl iodide (IV.) was formed, which on reduction with zinc
and hydrochloric acid gave suberane (V.).

CH2 . CH2 . CH2 . COOH CH2 . CH2 . CH2

CH2.CH2.CH2COOH CH2.CH2.CO

(I.) (II.)

CH2.CH2.CH2

> CH

CH2.CH2.CH.OH CH2.CH2.CH.I CH2.CH2.CH2

(IIL) (IV.) (V.)

It is a liquid, boiling at 118° C. under a pressure of 726 mm.
and having a peculiar odour like naphtha.

The preparation of octomethylene is much more difficult.
It has been carried out by Willstatter and Veraguth 2 in the
following steps. The bark of the "pomegranate tree contains
an alkaloid pseudo-pelletierine which is a ring-homologue of
tropinone, and has the constitution expressed by (I.). This was
converted into the N-methyl-granatanine (II.) in which two
hydrogen atoms replace the oxygen of the ketonic compound.

CH2 — OH 0x1.2 OH2 — OH OH2

III III

CH2 N— CH8 CO CH2 N— CH8 CH2

CH2— CH CH2 CH2— CH CH2

(I.) (II.)

By exhaustive methylation, this compound is converted
into des-dimethyl-granatanine (III.), the nitrogen bridge being
broken in the process.

1 Markownikoff, /. Rnss. Phys. Chem. Soc., 1893,25,364,547; Willstatter
and Kametaka, Per., 1908, 41, 14SO.

2 Willstatter and Veraguth, Ber., 1905, 38, 1975 ; 1907, 40, 957.

30 RECENT ADVANCES IN ORGANIC CHEMISTRY

N(CH3)3OH
— OH

CH2 CH2

I . I

Oii2 — OH — OH

(in.)

This substance, on distillation, breaks down into water, tri-
methylamine, and a cyclo-octadiene, CsH12. This last substance
is unstable, and polymerizes with almost explosive violence.
When hydrobromic acid is allowed to react with it, it forms a
dihydrobromide, C&HuBty, from which hydrobromic acid can be
removed by means of quinoline. The compound thus produced,
however, is not the original cyclo-octadiene, but an isomeric
and much more stable body, The constitution of neither
compound has been established with certainty, but, as will be
seen immediately, this does not affect the present question.
When the stable cyclo-octadiene is reduced by the Sabatier
and Senderens method it produces a hydrocarbon, CsHig,
boiling between 147° and 149°. This substance is not an
olefine, as it is unattacked by permanganate of potash ; so that
it must be a polymethylene of some sort. On oxidation with
nitric acid it produces suberic acid, which proves that all the
carbon atoms lie in a single straight chain —

CH2 . CH2 . CH2 . CH2 CH2 . CH2 . CH2 . COOH

I I -- > I

CH2 . CH2 . CH2 . CH2 CH2 . CH2 . CH2 . COOH

The only possible conclusion is that the hydrocarbon is
octomethylene. It is a solid, melting at 14'2° C. and boiling at
150° C. under 709 mm. pressure.

The last compound with which we have to deal is nono-
rnethylene, which was recently discovered by Zelinsky.1 He
obtained it by the second general method, starting from sebacic
acid (I.), which, by distillation of the calcium salt, he converted
into the corresponding ketone (II.), and thence, by reduction, to
the alcohol (III.), which, by conversion into the iodide and
further reduction, gave nonomethylene (IV.).
1 Zelinsky, Ber., 1907, 40, 3277.

THE POLYMETHYLENES

GH2 . GH2 . Grl2 . GH2 . COOH GH2 . GH2 . GtT2 . GH2

CH2.CH2.CH2.CH2
(II.)

CH2 . CH2 . CH2 . CH2 . COOH

(I.)
CH2 . CH2 . CH2 . CH2

^CH . OH | ^CH2

riTT r*ir PIT r<TT r<ir PIT PTT PIT

V^X12 . wXl2 . L^X12 . \jii-2 v^JLl2 . V^-H2 . \jtt.% . v^JLljj

(III.) (IV.)

It is a liquid of boiling-point 170°-172° C.

We have now given a sufficient account of the methods by
which these substances can be formed, and must next take up
the question of the effects which the ring-formation produces
upon the general type of polymethylenes. These saturated
cyclic compounds occupy a peculiar position in the field of
organic chemistry. Eelated on the one hand to the aromatic
series, from which some of them can be derived, they resemble
aromatic bodies to some extent in their stability ; while on
the other hand their actions with certain reagents bring them
more into line with the defines, whose isomers they are. In
stability they seem to mark a transition stage between the
ordinary define and the analogous saturated paraffin. In
physical properties also the polymethylenes lie apart from
both define and paraffin series ; and it may be well to examine
this part of the subject before dealing with the chemical
behaviour of the cyclic group.

From the point of view of chemistry the boiling and
melting points of a substance are two of its most important
physical properties, as by their aid we can separate or identify
isomeric compounds. We may, therefore, begin by considering
the boiling-points of the olefinic, polymethylene, and paraffin
derivatives, comparing in each case the three compounds which
have the same number of carbon atoms in the chain —

Boiling-point of

No. of

carbon atoms.

Olefinc.

Polymethylene.

Paraffin.

-48°

circa - 35°

-45°

- 5°

+ 12°

+ 1°

+ 40°

49°

36°

69°

81°

69°

95°

117°

98°

122°

146°

126°

—

171°

150°

32 RECENT ADVANCES IN ORGANIC CHEMISTRY

Thus in every case the boiling-point of the polymethylene
is the highest of the three. This emphasizes the peculiar cha-
racter which the ring-formation confers upon substances, for
in most cases the saturated (paraffin) compound has almost the
same boiling-point as the corresponding olefinic derivative.

In molecular volumes also1 the polymethylenes lie quite
apart from the olefines and paraffins, as the following table
shows : —

No of

Molecular volumes of

carbon atoms

in chain.

Olefine.

Polymethylene.

Paraffin.

89-8

79-06

96-5

104-3

91-09

112-4

119-1

105-19

127-2

136-3

118-00

142-5

151-5

130-92

158-6

~~~

159-46

174-3

From this it appears that the molecular structure of the
polymethylenes is much more compact than that of the corre-
sponding olefines; and, further, the higher polymethylenes
are relatively less voluminous than the lower members. For
example, the difference in volume between the two isomeric
compounds with four carbon atoms is ten units, while that
between the volumes of isomeric compounds of eight carbon
atoms is twenty units ; over the same interval the difference
between the olefine and corresponding paraffin remains almost
unaltered — seven units.

Briihl2 has shown that ring-formation has no noticeable
effect upon the molecular refraction of compounds; thus the
difference between the refractive power of a saturated paraffin
and that of the corresponding ring is to be found merely by
subtracting the value of two hydrogen atoms from the larger
figure, taking no account of the change in constitution.

Stohmann and Kleber 3 have examined the question of the
relation between ring-formation and thermo-chemical behaviour
in an exhaustive manner. In the following table column I.

1 Willstatter and Bruce, Ber., 1907, 40, 3979, and Smiles, Relation between
Chemical Constitution and Physical Properties, Chapter IV.

2 Briihl, Ber., 1892, 25, 1954; Willstatter and Bruce, Ber., 1907, 40, 3979.

3 Stohmann and Kleber, /. pr. Ch., 1892, II. 45, 489.

THE POLYMETHYLENES 33

shows the increase in the heat of combustion when a poly-
methylene ring is broken and two hydrogen atoms are added
on; column II. shows the average loss of energy in calories
which the polymethylene system suffers by the addition of two

hydrogen atoms : —

I. II.

Trimethylene ring . . . 31-9 33-1

Tetramethylene ring . . . 29'1 39-9

Pentamethylene ring . . 52-9 16-1

Hexamethylene ring . . . 54'7 14 -3

The only point of importance which can be deduced from these
figures is the fact that the penta- and hexa-methylene rings lose
much less energy in opening up than the tri- and tetra-methy-
lene ones do. We shall have occasion to refer to this point
later in the chapter.

When we come to the chemical side of the question, the
evidence is not nearly so complete as is desirable. A good
deal of research has been carried out on the problem of the
stability of polymethylenes in presence of such agents as
halogen acids, permanganate, nitric, and sulphuric acid ; but
up to the present no one appears to have done any exact
comparative experiments which would enable us to consider
numerical relations between the different cases. We must,
therefore, content ourselves for the present with noting the
main features of the matter.

Trimethylene is comparatively unstable. It is attacked by
the halogen acids and by sulphuric acid, the ring being opened
in each case. Potassium permanganate acts on it slowly, which
distinguishes it from propylene, the latter being instantly
oxidized. Berthelot1 gives the following data of comparison
between the olefine and polymethylene : —

Formation.

Heat iu calories of
Bromine addition.

Sulphuric acid
addition.

Trimethylene
Propylene .

-17-1
- 9-4

+ 38'5
4-29-1

+ 25-5
+ 16-7

From this it appears that trimethylene has an energy-con-
tent eight calories greater than that of propylene.

1 Berthelot, C. R., 1899, 129, 483.

34 RECENT ADVANCES IN ORGANIC CHEMISTRY

When we turn to the next higher member of the series,
tetramethylene, we find that it is more stable. It is not
attacked by cold concentrated hydriodic acid or by bromine in
chloroform solution. When passed through a heated tube in
the Sabatier and Senderens method it requires a temperature
of about 200° to reduce it to butane,1 whereas trimethylene is
changed to propylene at 100° 2 by passing it over heated iron.
These last two data are not quite comparable, but certainly
point to trimethylene being less stable than tetramethylene.

Pentamethylene is a stable substance, being unattacked by
hydriodic acid even when boiling.

Hexamethylene appears to be as stable as pentamethylene.
It is attacked by chlorine, but instead of the ring being opened,
substitution takes place.

The almost equal stability of the penta- and hexa-methylene
systems is well shown by a peculiar series of changes by which
hexamethylene derivatives can be isomerized into pentamethy-
lene compounds, and vice versa. Many such changes are known,
and for the sake of illustration we may quote one or two here.

Aschan3 has shown that when hexamethylene is treated
with anhydrous aluminium chloride it is converted below 100°
into methyl-pentamethylene. The change appears to be a purely
desmotropic one, for no discoloration of the liquid was observed,
nor were any bye-products of condensation found, such as were
to be expected if the hexamethylene ring had been broken.
Perkin and Yates * found that when camphoric anhydride was
treated with aluminium chloride, hexahydro-xylylic acid was
formed —

COOH CH3 H

\/
C

OH — O
CH— COOH

1 Willstatter and Bruce, Per., 1907, 40, 3979.

2 Ipatjeff, Ber., 1902, 35, 1063.

3 Aschan. Annale-n, 1902, 324, 11.

4 Perkin and Yates, Trans. Ghent. Soc., 1900, 79, 1373 ; Lees and Perkin,
ibid., 1901, 79, 332.

THE POLYMETHYLENES 35

This tends to show that the five- and six-membered rings are
of almost equal stability.

The ring in the next homologue of the series, heptamethy-
lene, is less stable than either the five- or six-membered
substances. Markownikoff * has observed that when iodo-
heptamethylene is heated with hydriodic acid to 250° it is
converted into methyl-hexamethylene and dimethyl-pen ta-
methylene —

CH.CHs CH.CH3

7 XCH HCXNCH HCXXCH

r^TT rim TI o riTT IT r* i

UHo OXlo -H-oV v^-H.2 tl<Aj (

H2C — CH2 — CH2 H2C CHg tl^C CH . CHs

CH2

With regard to the behaviour of octomethylene and nono-
methylene, the experimental data at our disposal are too scanty
to allow of any but very general conclusions being drawn with
regard to their stability. It appears that they are less stable
than the five- and six-membered rings, but no exact measure-
ments have been made.

Enough has now been said to prove that the polymethylenes
show somewhat peculiar relations between their stabilities and
the number of carbon atoms in the ring. The five- and six-
membered rings are the most stable, and the stability decreases
from this maximum, whether the number of carbon atoms in
the ring be increased or diminished. Thus, if we take tri-
methylene and increase the size of the ring by a methylene
group, we obtain the more stable tetramethylene ; a further
introduction of a methylene group yields a further increase in
stability, pentamethylene being formed. The next methylene
group, leading to hexamethylene, hardly affects the stability;
but any further inclusion of methylene radicals, instead of
increasing the stability as before, tends now in the opposite
direction, hepta-, octo-, and nonomethylene being each in turn
less stable than its lower homologue.

Taking this into account, Baeyer2 put forward what is

1 Markownikoff, Ber., 1897, 30, 1214.

2 Baeyer, Ber., 1885, 18, 2277.

36 RECENT ADVANCES IN ORGANIC CHEMISTRY

generally known as his " Strain Theory," which may be formu-
lated as follows : —

" The four valencies of a carbon atom act parallel to the lines joining the
corners of a regular tetrahedron to its centre, making an angle of 109° 28'
with each other. The direction of the valencies can be altered, but any
such alteration produces a strain whose amount is proportional to the angle
through which the valencies are diverted.'1

On Baeyer's view, the state of strain in the ring is a
measure of the ring's stability ; the greater the strain the less
stable the ring is likely to be. We must now apply this theory
to the seven polymethylenes, and see how far it agrees with
experimental results.

In trimethylene the centres of three carbon atoms will lie
at the corners of an equilateral triangle, so that the valencies
joining these carbon atoms to each other will make an angle
of 60° with each other. But in the original state of things
these valencies were supposed by Baeyer to be inclined to
each other at an angle of 109° 28' ; so that two of the valencies
of any carbon atom have been diverted through an angle of
(109° 28' - 60°), and each valency has been diverted through
half this angle, viz. 24° 44'. Similarly, in the case of a tetra-
methylene ring, the four carbon atoms lie at the corners of
a square, and the valencies will make an angle of 90° with
each other. Thus, each pair of valencies has been diverted
through an angle of (109° 28' - 90°), and each single valency
has been turned through half this angle, viz. 9° 44'. The
general formula giving the deviation for a ring-compound
containing n carbon atoms is —

Applying this to the seven polymethylenes, we obtain the
following values for the deviation in each case : —

Angle of deviation.
(Ethylene) . » , v . 54° 44'

Trimethylene

Tetramethylene

Pentamethylene

Hexamethylene

Heptamethylene

Octomethylene

Nonomethylene

24° 44'
9° 44'
0°44'

- 5° 16'

- 9° 33'

- 12° 46'

- 15° 16'

THE POLYMETHYLENES 37

These results are in moderate agreement with the actual
relations between the stabilities of the polymethylenes. The
positive deviations are more in accordance with experimental
results than the negative ones. Of course it must be under-
stood that in no case could the Strain Theory pretend to
exactitude, since it assumes that the four valencies in the
grouping—

H 0

v • .

/\
H C

are evenly distributed in space, which is most unlikely.
Apart from this, however, the agreement between the theory
and the facts is noteworthy ; and it is very desirable that we
should have more exact data at our disposal with regard to
the stability relations of these compounds, in order to discover,
if possible, what the Strain Theory actually corresponds to in
physico-chemical relations.1

We have now completed our survey of the polymethylenes,
and in the next chapter we shall examine the derivatives of
the simple ring compounds which occur among the terpenes.

1 Compare Smiles, Relations between Chemical Constitution and Physical
Properties, 1910, pp. 268 et eeq.

CHAPTER III

THE MONO-CYCLIC TERPENES

1. Introductory

WHEN the saps and tissues of certain plants (such as pines,
camphor, lemons, and thyme) are distilled, the distillates are
found to contain among other things a mixture of substances
which are classed under the general head of ethereal oils. For
the most part these ethereal oils contain unsaturated hydro-
carbons of the general formula (CgHs),, (or derivatives of these
substances), and these may be divided into three classes —

1. Open-chain olefinic compounds.

2. Mono-cyclic hydrocarbons (reduced benzene derivatives).

3. Cyclic compounds containing more than one ring.

In the following chapters we shall consider the first and
third of these classes, while the present chapter will be devoted
to the mono-cyclic substances.

In the naturally occurring compounds it is found that
by far the greater number of these hydrocarbons have the
empirical formula CioHi6 ; and it is not without interest that
Collie,1 in polymerizing ethylene by means of the silent electric
discharge, found that the major part of the substance used was
converted into compounds containing either ten or fifteen
carbon atoms.

The nomenclature of these substances is at present some-

what in confusion. It has been customary to apply the name
terpene to any compound having the composition C5Hs, or any
polymeric variety of this type. This general type was then
divided into two others : the " true terpenes," cyclic substances
of the formula Ci0Hi6 ; and the " olefinic terpenes/' which
are open- chain bodies having the formulse CsHs and CioHio.
Another system of nomenclature classes the whole group under

1 Collie, Tram. Ghem. Soc., 1905, 87, 1540.

THE MONO-CYCLIC TERPENES 39

three heads : hemi-terpenes, C5H8 ; terpenes, Cl0Hi6 ; and sesqui-
terpenes, C15H24. It will best serve our purpose to divide the
terpenes into the three classes which we mentioned first, viz.
olefinic terpenes, mono-cyclic terpenes, and dicyclic terpenes.
The naturally occurring mono-cyclic terpenes are for the most
part derived from either m- or ^>-hexahydrocymene.

Most of the terpenes are colourless, pleasant-smelling liquids
of high refractive power. They boil without decomposition,
and are volatile in steam. Some are optically active, some
inactive by racemization, while others, containing no asymmetric
carbon atom, cannot show activity at all. It is not necessary
to deal with their chemical properties at present, as these will
be brought out in the following pages when the constitutions
of the compounds are described.

2. The Synthesis of Terpineol.

In the group of the mono-cyclic terpenes, by far the most
important compound is terpineol, for from it most of the
other members of the group can be prepared, either directly
or indirectly. The constitution of terpineol, therefore, is of
considerable value to us in determining the constitutions of
other substances which we can derive from it. The inactive
form of terpineol has been synthesized by Perkin,1 and as this
synthesis determines the constitution of the substance, we may
describe it step by step.

When /3-iodo-propionic ester was allowed to interact with
the disodium derivative of cyan-acetic ester, -y-cyano-pentane-
ayt-tricarboxylic ester was produced —

ON EtOOC.CH2.CH2 ON
2EtOOC.CH2.CH2 1 + Na2C = 2NaI + G

COOEt EtOOO.CH2CH2 COOEt

From this the free acid was obtained by hydrolysis with
hydrochloric acid, and when it was boiled with acetic
anhydride and then distilled it was transformed by loss of
water and carbon dioxide into S-keto-hexahydrobenzoic acid —

1 Perkin, Trans. CJirm. Soc., 1904, 85, 654,

40 RECENT ADVANCES IN ORGANIC CHEMISTRY
HOOC . CH2 . CH2

CH . COOH = H20 + C02 +

HOOC.CH2.CH2

CH2.CH2

+ CO CH.COOH

CH2 . CH2

»•

Grignard's reaction was then applied to the ester of this acid,
magnesium methyl iodide being allowed to react with the
ketonic group, and in this way S-hydroxy-hexahydrotoluic
ester was formed —

CH3 CH2.CH2

\ / \
Mg + 00 CH . COOEt >

/ \ /

I CH2.CH2

CH3 CH2.CH2

\ H2o
> 0 CH. COOEt >

/ \ /

IMgO CH2.CH2

CH3 CH2 . CH2
H20 \ / ' \

> C CH. COOEt

.CHo

HO CH2.CH2

When, by the action of fuming hydrobromic acid, we replace
the hydroxyl group in this acid by a bromine atom and
then remove hydrobromic acid from the compound by means
of weak alkalis or pyridine, we obtain A3-tetrahydro-j9-toluic
acid —

CH3 CH2.CH2

C CH. COOEt >

HO CH2.CH2

THE MONO-CYCLIC TERPENES 41

HBr \

-- > C OH . COOH

\
CH2 .

Br CH2 . CH2

CH3 . .

Alkalis \ / \

> 0 CH.COOH

V C/

Oxl . U±±2

After esterifying the acid, the Grignard reaction can be again
employed, with the result that the ester group is attacked, and
on treatment with water the intermediate compound breaks
down into inactive terpineol.

CH2.CH2 CH2.CH2 CH3

/ \ CHg.Mg.I / /

CH3.C CH.COOEt > CH3.C .CH-C-OH

CH . CH2 OH . CH2 CH3

Terpineol.

If this synthesis be examined step by step it will be seen
that there can be no doubt as to the constitution of terpineol,
for the reactions can only be supposed to take place in the way
shown. Any alternative formulation of any of the reactions
would at once lead to contradiction in the later experiments.

An optically active terpineol has been prepared by Fisher
and Perkin 1 by resolving the intermediate acid into dextro and
Isevo forms before continuing the synthesis.

3. The Decomposition Products of Terpineol.

The oxidation of terpineol takes place in several steps and
produces some compounds of importance in the study of
terpene constitutions ; we may, therefore, deal with the matter
briefly in this place.

It has been shown by Wagner 2 that when a compound con-
taining a double bond is oxidized by means of potassium
permanganate, the first step in the process is the breaking of
the double bond and the addition of a hydroxyl group to each
of the atoms between which the double bond originally lay —

1 Fisher and Perkin, Trans. Chem. Soc., 1908, 93, 1871.

2 Wagner, Ber., 1888, 21, 1230, 3359; 1891, 24, 683.

42 RECENT ADVANCES IN ORGANIC CHEMISTRY

R— C— R R— C— R

|| + H20 + 0 = |
R_C— R R_C— R

In the case of terpineol this rule holds, and it is found that the
first oxidation product 1 obtained by the action of permanganate
upon terpineol is trihydroxy-hexahydrocymene —

CH3

C— OH

/ \

CH2 CH

CH2 CHOH

CH2 CH2

> 1 1

r^TT f ''TT

vyXi2 v>/jn2

\ /

C.OH

CH3 CH3

Terpineol.

Trihydroxyhexahydrocymene.

This substance, on further oxidation,2 is converted into homo-
terpenylic methyl ketone by the rupture of the single bond
between the two hydroxyl-bearing carbon atoms—

CH- CHa

CH2 COOH CH2 CO—

I I > I I

CH2 CH2 CH2 CH2

C—OH C— OH C 0

CH3 CH3 CH3 CH3 CH3 CH3

Trihydroxyhexa- Intermediate acid. Homoterpenylic

hydrocymene. methyl ketone.

1 Wallach, Annalen, 1893, 275, 150.

2 Ibid.; Ber., 1895, 28, 1773; Tiemann and Schmidt, {bid., 1781.

THE MONO-CYCLIC TERPENES

As is shown in the formulae, the first product of the oxida-
tion is a hydro xy acid which loses water at once between its
carboxyl and hydroxyl groups, yielding the keto-lactone.
This keto-lactone is the first product which can be isolated
when terpineol is oxidized with chromic acid, for the action is
so violent that the trihydroxyhexahydrocymene is destroyed
as soon as it is formed.

Further oxidation with potassium permanganate l converts
the keto-lactone into a mixture of acetic and terpenylic acids —

CH3

H2C CO—

H9C

OIL

C—

CH3
CO. OH

COOH CO—

I . I
CH2 CH2

-0

CH3

Homoterpenylic methyl ketone.

CH3

. CH8
Terpenylic acid.

The latter substance, by the action of a five per cent, solution
of permanganate, is still further decomposed into terebic acid —

COOH CO-

I |,

CH2 CH2

"CH

COOH

\ „
CH

CO—

OH2

-0

CH3

Terpenylic acid.

CHa

Terebic acid.

It will be seen that these formulae for homoterpenylic,
terpenylic, and terebic acid illustrate the decomposition of
terpineol quite satisfactorily. Any doubt as to their accuracy
was removed by the synthesis of the three acids, which was

1 Wallach, Ber., 1895, 28, 1776.

44 RECENT ADVANCES IN ORGANIC CHEMISTRY

carried out by Simonsen.1 Terebic 2 and terpenylic acid 3 had
previously been synthesized in different ways. The Siinonsen
syntheses depend on the application of Grignard's reaction to
various ketonic esters. From magnesium methyl iodide and
acetyl-succinic ester he obtained terebic ester —

COOEt
COOEt CH2

"CH

COOEt

CH3 CH3

Terebic ester.

CH3

Acetyl-succinic ester.

In exactly the same way /3-acetyl-glutaric ester is converted
into terpenylic ester, and [3-acetyl-adipic ester into homo-
terpenylic ester.

The constitution of terpineol, then, may be considered to be
completely established, both synthesis and degradation products
agreeing with the theory.

4. The Constitution of Dipentene.

When terpineol is heated with acid potassium sulphate it
loses a molecule of water, and is converted into dipentene. It
is evident that we may represent this elimination of water in
either of two ways —

CH3 CH3 CH3

; i . ' A .'.•: A

HC CH

H2C CH

C.OH

CH,

CH3 CH3

(I.) Terpineol.

Simonsen, Trans. Chem. Soc., 1907, 91, 184.

'H.q

3 CH2

(II.)

Blaise, C. R., 1898, 126, 349.

Lawrence, Trans. Chem. Soc., 1899, 75, 531.

THE MONO-CYCLIC TERPENES 45

Now, dipentene can be obtained by mixing together equal
quantities of dextro- and Isevo-limonene. It is, therefore, the
racemic form of limonene, and must contain an asymmetric
carbon atom. Formula (I.) contains no such carbon atom,
but the atom in (II.), which is marked with an asterisk, is
asymmetric. Dipentene, then, must have the constitution
represented by (II.).

In order to satisfy ourselves that this formula is the
correct one, we may test it by seeing how far it agrees
with some decompositions which dipentene can be made to
undergo.

When nitrosyl chloride is allowed to act upon a compound
containing a double bond it may unite with it in either of two
ways.1 If the double bond lies between two tertiary carbon
atoms, the chlorine atom attaches itself to the one and the
nitroso-group to the other, and the resulting substance is a blue
nitroso-derivative —

C=C NOCI C--C

CHs CH3 CHg

NO 01

On the other hand, if one of the carbon atoms is a tertiary and
the other a secondary one, the chlorine of the nitrosyl chloride
attaches itself to the tertiary atom and the nitroso-group to the
secondary atom. The hydrogen atom then wanders, as shown
in the formulae below, with the result that a colourless iso-mtroso
compound is formed —

CH3 CH3 CII

0=0 NOCI C— 0 0—0

/ \ . -- > A ,\

CH3 H CH3! H OJ

01 NO 01 NOH

We must now apply this to the case of dipentene. To make
reference easy we shall number each step.

1 Thiele, Ber., 1894, 27, 455..

46 RECENT ADVANCES IN ORGANIC CHEMISTRY

I. When nitrosyl chloride acts upon dipentene, it might be
supposed that it could react either with the double bond in the
nucleus or with that in the side-chain. It actually attacks
the nuclear double bond, as we shall show later, and to avoid the
complication of two sets of formulae we may confine ourselves
to the case of the addition to the double bond of the nucleus.
The reaction, if our formula for dipentene be correct, will take
the course shown below —

CH3
I

C C_C1

/\ /\

H2C CH NOCI H2C CH:NO

I I > I I

H2C CH2 HoC

CH3 CH<

Dipeotene.

CH3

C— Cl

H2C C : NOH

I I
H2C CH2

\/
CH

CH3 CH2

II. When the nitrosochloride formed in the last reaction is
treated with alcoholic potash it loses one molecule of hydro-
chloric acid, and is transformed into a compound which proves
to be identical with the oxime of the ketone carvone. This can
be expressed as follows : —

CH3

C— Cl

H2C C : NOH
H2C CH2

HCX C : NOH

H2C CH2

CH2

Dipentene nitrosochloride.

CH3 CH2
Carvoxime.

THE MONO-CYCLIC TERPENES 47

III. By hydrolysis of the oxime, carvone is produced.

IV. Carvone, on reduction, gives dihydro-carveol. This
reduction might be supposed to take place either in the nucleus
or in the side-chain. As will be shown later (VI.), the nucleus
is reduced and the side-chain left untouched. We need not
concern ourselves with the alternative set of formulae, but may
again confine ourselves to the one set.

CH3

I I

C CH

HC CO H2C CH.OH

| | > I

HP PTT TI C*

v\j v^llo xlavy

i i

Carvone. Dihydrocarveol.

V. On oxidation, dihydrocarveol gives a trihydroxy-hexa-
hydrocymene —

CH3

H2C CH.OH

H2C CH3

C— OH
CH3 CH2OH

VI. On further oxidation a ketone alcohol is formed —

48 RECENT ADVANCES IN ORGANIC CHEMISTRY

CH3

H2C CH . OH

H2C CH2

I
CO

CH,

The production of this substance proves what was previously
stated in I. and IV., viz. that the nitrosyl chloride attacks
the nucleus, and that in the reduction to dihydrocarveol the
side-chain double bond is not reduced. If the nitrosyl chloride
had attacked the side-chain we should, at Stage III., have
produced an aldehyde of the type —

CH3

H2C CH

i I

H2C CH2

CH2 CHO

instead of the ketone produced in practice. If the side -chain
had been reduced in Stage IV. instead of the nucleus, the
nucleus would have been attacked by the oxidizing agent in
Stage V., the ring would have been broken, and a ketonic acid
would have been formed, just as in the case of the oxidation of
terpineol.

VII. Further oxidation of the ketonic alcohol produced in
Stage VI. yields a hydroxy-acid, which, by the action of
bromine at 190° C., loses six hydrogen atoms, and is converted
into hydroxy-p-toluic acid —

THE MONO-CYCLIC TERPENES 49

CH,

H2C CH.OH

Hatf

CH.OH

HC COH

i i

' H2C

HC CH

H2C CH2

CH2

\< V

Oxl CM

CO COOH C

OOH

CH3

To sum up the matter, we may point out that the series
of reactions IV. to VII. prove that the "isopropyl group"
contains a double bond, which must also be present in dipentene.
Moreover, since this double bond has persisted throughout the
whole series of reactions I. to IV., it cannot have been the point
at which the nitrosyl chloride attached itself, as this portion
of the molecule has given rise to the — CH . OH — group.
Further, the nitroso-group must have attached itself to the
carbon atom to which the hydroxyl group is attached in the
aromatic acid, i.e. the one next that which carries the methyl
group. These reactions can only be explained by assuming that
dipentene has the structure which we attributed to it on account
of its synthesis from terpineol.

It might be objected that we have not taken into account
the possibility that, in the formation of dipentene, the elimi-
nation of water from terpineol may take place between two
non-adjacent carbon atoms, giving rise to some such compound

50 RECENT ADVANCES IN ORGANIC CHEMISTRY

Any attempt to explain the question on these lines leads,
however, to impossible results, and it may be taken as proved
beyond doubt by the above Q jj

experimental data that the |

formula of dipentene is — C

H2C CH

H2C

\/
CH

CHS

Dipentene.

This, in turn, proves the formulae of dextro- and laevo-limonene,
for as they are the optical antipodes of which dipentene is the
racemic variety, they also must possess the same structural
formula as dipentene.

5. The Constitutions of Terpinolene and Terpinene.
In the last section it was pointed out that the dehydration
of terpineol might follow either of two courses : the one leading
to a compound containing an asymmetric carbon atom, the other
to a symmetrical derivative. The result of dehydration by
means of acid potassium sulphate was shown to be dipentene ;
but when terpineol is dehydrated1 by means of alcoholic
sulphuric acid, an QH3

isomeric compound is J

formed which has

the second of the two / OH

possible formulae — ,

HgC CHg

^ 2S A 1

Cos CH/j
This substance is terpinolene.

Now, according to Thiele,2 the grouping (I.) is less stable

1 Wallach, Ber., 1879, 12, 1022. 2 Thiele, Annalen, 1899, 306, 119.

THE MONO-CYCLIC TERPENES 51

than the grouping (II.) in which the two double bonds are
" conjugated " —

EREEE EEEEE

I I I I I I I I I I

C = C— C— C = C E— C— C = C— C = C

III I I

E E E E E

(I-) (II.)

We need not enter into the matter in detail here, as it will be
dealt with fully in a later chapter. For the present it is
sufficient to apply Thiele's view to the behaviour of terpinolene.
This substance, on treatment with acids, can be converted into
terpinene, while terpinene itself cannot be isomerized at all,
and is, in fact, the most stable of all the terpene class. Since
the grouping (I.) exists in terpinolene, we may conclude that it
is converted by acids into the more stable grouping (II.) —

CH3 CH3

H2C CH HC CH

.HO Clr

C C

II I

C CH

CH3 CH3 CH3 CH3

Terpinolene. Terpinene.

This, however, is only a possibility and not a certainty, for the
alternative formula
of terpinene —

2C
HC C

H2C CH
H2

I
CH

CHa CH

52 RECENT ADVANCES IN ORGANIC CHEMISTRY

represents some properties of the substance better, though it
does not explain its stability so well. The matter is still under
discussion, and need not be dealt with further in this place.

6. Terpin and Cineol.

Grignard * and others have shown that when the esters of
organic acids react with organo-magnesium compounds, tertiary
alcohols can be produced — R

2E.Mg.I + E'.COOEt = E'.C— OMg. I + EtO.Mg.I

E
K E

E' . C— 0 . Mg . I + H20 = E' . C— OH + HO . Mg . I

E E

Again, when ketones are treated with Grignard's reagent,2
tertiary alcohols are formed —

E E E

R'.Mg.I H20

CO > E'— C— 0.

CO > E'—C— O.Mg.I > E'— C— OH

E E E

Kay and Perkin 3 have combined these two reactions into
one, using a ketonic ester, and allowing both vulnerable groups
to be attacked simultaneously. By this means, from cyclo-
hexanone-4-carboxylic ester, they obtained the dihydric alcohol
terpin. CHa

CO C— OH

r/\, /S

TT /^1 r^TT TT C\ C\~r~T

±i2^ on2 J120 CH2

II II

H2C CH2 H2C CHo

)OOEt C— OH

Cyclohexanone-4-carboxylic ester. / \

CH3 CH3

Terpin.

Grignard, C. R., 1901, 132, 336. 2 Zelinsky, Ber., 1901, 36, 3950.

3 Kay and Perkin, Trans. Ghent. Soc., 1907, 91, 372.

THE MONO-CYCLIC TERPENES 53

This synthesis proves the formula of terpin beyond any
dispute.

Terpin may be also obtained by boiling terpineol with dilute
sulphuric acid —

CH3 CHo

C— OH

/^ . /\ .

-ti2C Cxi H2C CH2

I l( H.O | I

H2C CH2 ^ H2C CH2

CH H— C

! I

C— OH C— OH

GH3 CH3 CH3 CH3

Terpineol. Cis-terpin.

The terpin which is obtained in either of these ways is
called a's-terpin, from the fact that in its space formula the two
hydroxyl groups lie on the same side of the hexamethylene
ring, while in the isomeric compound, £nms-terpin, they lie
on opposite sides of the ring —

OH HO— C(CH3)2 CH3 HO— C(CH3)2

CH2 - CH2, |

|
C

|
O

CH3 H OH H

Cis-terpin. Trans-terpin.

Cis-terpin unites with one molecule of water to form terpin
hydrate, a crystalline substance from which it can be regene-
rated at 100° C. The trans-isomer does not unite with water
at all.

Cis-terpin cannot be directly converted into trans- terpin,
but the change can be effected by a somewhat roundabout
method. In the first place, cis-terpin is subjected to the action
of hydrobromic acid, by which means a dibromide is formed.
As can be seen from its formula, this substance is identical with
the hydrobromide of dipentene —

54 RECENT ADVANCES IN ORGANIC CHEMISTRY

CH3
C— Br

H2w OJtlg

H2C CH2

H— C

* C— Br
CH3 CH3

Dipentene dihydrobromide
(Cis-terpin dibromide).

This dibromide is next treated with silver acetate in acetic
acid solution, and the diacetate so produced is hydrolyzed with
alcoholic potash, yielding tfnms-terpin.

CH3 OH3

Br— C CHaCO . 0— C C— OH

H2C CH2 H2C CH2 H.2C CH2

I I -- > I I -- > I I

H2C CH2 H2C CH2 H2C CH2

C— H G— H C— H

Br— C CHgCO . 00 HO— C

/\ /\ /\

CH3 OH3 CH3 CH3 CH3 OH,

It should be noted that when cis-terpin is converted into its
dibromide the product is the cts-form of dipentene dihydrobro-
mide ; while, on the other hand, the action of hydrobromic acid
on trans-terpin produces the trans- variety of dipentene dihydro-
bromide. Thus the change of cis-terpin into trans-terpin
cannot be carried out through the bromides alone, as during
their formation no change from cis- to trans-form takes place ;
this only occurs during the hydrolysis of the acetyl derivative.

When cis-terpin is dehydrated, it yields a variety of

THE MONO-CYCLIC TERPENES

products (terpineol, dipentene, terpinene, and terpinolene),
among which is found the compound cineol, Ci0H180. This
substance contains neither a hydroxyl nor a carbonyl radical,
and must therefore be an ether. On this view, its formation
from cis-terpin is easily explained —

CH3— C— OH

/\
H2C CH2

H2C CH2

H— 0

C-OH
CH3 CH3

CH3— C-

H2C CH2

H2C CH2 0

H— C

'Hq

CH3

This formula is supported by the fact that hydrobromic acid
in acetic acid solution converts cineol into cis -dipentene
dibromide —

CHa— C-

H2C CH2 0

CH3

CH3— C— Br
HC/NCH

Xi2v>' V^Jn.2

H—

C— Br
CH3 CH3

The behaviour of cineol on oxidation with potassium per-
manganate is curious.1 The first effect is to break the hexa-
methylene ring, while leaving the ether chain untouched ; in
this way cineolic acid is produced —

1 Wallaeli and Gildemeister, Annalen, 1888, 246, 268 ; Wallach, ibid., 1890,
258, 319; Wallach and Elkeles, ibid., 1892, 271, 21.

CH3
-C
H2C CH2

56 RECENT ADVANCES IN ORGANIC CHEMISTRY

CHg

H2C COOH
0 H2C COOH

CH3 CH3

Cineolic acid.

0 H2C CH2

0 O £

Cineol.

When cineolic acid is treated with acetic anhydride it yields
cineolic anhydride, which, on dry distillation, breaks down
quantitatively into carbon monoxide, carbon dioxide, and
rnethyl-heptenone, an aliphatic ketone of considerable interest
from its relations to the terpenes —

CH3
CO

CH2
CH2

C02

CH3 CH3

Cineolic anhydride.

)H3 CH3

Methyl heptenone.

7. The Synthesis of Carvestrene.

Until quite recently, carvestrene could be obtained only by
a very long and complicated series of reactions ; and the con-
stitutions of some of the intermediate compounds produced had

THE MONO-CYCLIC TERPENES

not been well established. Perkin and Tattersall1 have now
succeeded in synthesizing it by a series of reactions analogous
to those employed by Perkin in his synthesis of terpineol.

The starting-point of this new synthesis was ra-hydroxy-
benzoic acid. This was first reduced with sodium and alcohol,
forming hexahydro-m-hydroxy-benzoic acid ; from which, by
oxidation with chromic acid, y-keto-hexahydrobenzoic acid (I.)
was obtained. The ester of this acid reacts with magnesium
methyl iodide, giving the lactone of y-hydroxy-hexahydro-ra-
toluic acid (II.). When this is heated with hydrobromic acid
it yields -y-bromohexahydro-m-toluic acid (III.), which on treat-
ment with pyridine loses hydrobromic acid, and is changed
into tetrahydro-w-toluic acid (IV.). After esterification, this is
treated with magnesium methyl iodide and water, whereby an
alcohol (V.) is produced which differs from terpineol in that the
hydroxyl and methyl groups are in the 1, 3 position to each
other, while in terpineol they are in the 1, 4 position. Just as
terpineol, when treated with acid potassium sulphate, loses
water to form dipentene, this new alcohol loses water and forms
carvestrene (VI.).

CH3

H / XCH

tt%\j v^£l2

H2C CH . COOH

H.CO

(i.)

(ii.)

HO7

JUavy

CH3
C— Br
CHS

H2C CH.COOH

\/
CH2

(III.)

H2C CH.COOH

CH2

(IV.)

Perkin and Tattersall, Tram. Chem. Soe., 1907, 91, 480.

58 RECENT ADVANCES IN ORGANIC CHEMISTRY

i i

H2C OH CH-3 ji2C

I i / I I

H2C CH.C— OH H2C CH.C

\/ \ \ /

CH2 CHs CH2

(VI.)

Though since the discovery of this new synthesis the old
way of preparing carvestrene has lost its value as a practical
method, we may give a very brief description of it here on
account of one transition which occurs in the course of the
reactions. The starting-point for the old synthesis was the
substance carvone, which we have already encountered. Now,
as can be seen from the formulae of the two substances, to
convert carvone into carvestrene we must shift the isopropylene
group from one carbon atom to the adjacent one. How this is
done will be seen in due course.

CH3

C CH3

HC CO C

H2C CH2 H2C CH CH2

•\/ I I /

CH H2C CH— C

\x \

r\-rr OTT

U±12 ^-tls

v, Carvestrene.

CH2

Carvone.

Carvone is first reduced with zinc dust and alcoholic potash
to dihydro-carvone ; hydrobromic acid is then added on, giving
dihydrocarvone hydrobromide * —

* When a halogen acid is added on to the double bond of an unsaturated
substance, the negative part (i.e. the halogen atom) always unites with that car-
bon atom to which the fewest hydrogen atoms are attached. For example, in the

THE MONO-CYCLIC TERPENES 59

CHg

H2C CO

H2C CH,

C— Br

CH3 CH3

Now, when this substance is treated with alcoholic potash it
gives up hydrobromic acid, but instead of regenerating a carvone
derivative it yields a new ketone, carone. Since on oxidation
carone yields 1, l-dimethyl-2, 3-trimethylene dicarboxylic acid
(caronic acid), it must contain a trimethylene ring. The
simplest way in which this can be explained is to assume that
carone has either of the formulas (I.) and (II.).

CHg CHg

CH CH

/\ /\

H2C CO H2C CO

H20 OH HO OH2

C CH8— C

CH3 CH3

ax.)

case given below the compound formed by the addition of hydrobromic acid to
(I.) is (II.), and not (III.)

CH, CH3 Br CH3

\ \l \

C = CH2 C— CH3 CH— CH2Br

CH3 CH3 CH3

(i.) (n.) (in.)

This is called the " Markownikolf Eule" (Her., 1869, 2, 660; Annalen, 1870,
153, 256).

60 RECENT ADVANCES IN ORGANIC CHEMISTRY

The first of these formulae is the one usually ascribed to
carone. We cannot enter into the details of the evidence here.

When carone is allowed to react with hydroxylamine it
forms the substance carone oxime, which, on reduction, produces
the amino-compound carylamine —

H2C

CH CH

C:NOH H2C CH.NH2

H2C CH H2C CH

CH C(CH3)2 CH C(CH3)2

Carone oxime. Carylamine.

When this body is treated with alcoholic acid it undergoes
isomeric change, and is converted into the hydrochloride of
vestrylamine, the trimethylene ring being now broken. By
this means we have transferred the isopropylene group from
one carbon atom to the other —

CH3 CH3

CH CH

H2C CH.NH2 HCI H2C CH.NH2

CH C(CH3)2 H2C CH— C

Carylamine. \ /

Vestrylamine.

Vestrylamine hydrochloride, on dry distillation, breaks down
into carvestrene by loss of ammonium chloride —

C10H17 . NH2 . HCI = C10H16 + NH4C1

Carvestrene is a racemic compound, the dextro-antipode of
which is found in nature as sylvestrene.1 The latter has
recently been synthesized by Perkin.2

1 Baeyer, Ber., 1894, 27, 3485.

2 Perkin, Proc., 1910, 26, 97.

THE MONO-CYCLIC TERPENES 61

8. The Synthesis of Mentkone.

Though menthone had been synthesized in different ways
by Einhorn and Klages,1 Kotz and Hesse2 and Haller and
Martine,3 none of these methods furnished any proof of the
constitution of the substance. It was not until 1907 that
synthetic evidence was obtained upon this point.

Kotz and Schwarz 4 first synthesized /3-methyl-a'-isopropyl-
pimelic acid, and by the distillation of its calcium salt they
produced menthone —

CH3 CH3

H2C CH2.COO H2C CH2

I I' I I

H2C COO - Ca H2C CO

Calcium £-Methyl-a'-isopropyl-pimelate. Menthone.

A similar result is obtained by making the ester of this
acid undergo intramolecular acetoacetic ester condensation by
means of sodium, and then hydrolyzing the ester thus obtained
and splitting off carbon dioxide in the usual way.

CH3 CH3

CH CH

H2C CH2.COOEt H2C CH.COOEt

H2C COOEt H2C CO

v v

C3H7

1 Einhorn and Klages, Ber., 1901, 34, 3793.

2 Kotz and Hesse, Annalen, 1905, 342, 306.

3 Haller and Marline, C. 22., 1905, 140, 130.

4 Kotz and Schwarz, Annalen, 1907, 357, 206.

62 RECENT ADVANCES IN ORGANIC CHEMISTRY

CH-3

CH CH

H2C CH.COOH H20 CH

H2C CO H2C CO

CeHj 03X17

By means of this synthetic method, Kotz and Schwarz have
produced an active menthone which is strongly dextro-rotatory.

9. The Decompositions of Menthone.

Before the discovery of the syntheses which we have just
described, it had not been possible to show synthetically that
the methyl and iso-propyl radicals in menthone lay in the
para-position to each other. The evidence for this had,
however, been obtained from the decomposition reactions of
menthone.

When menthone is oxidized by means of potassium perman-
ganate, the first product is hydroxy-menthylic acid,1 which, on
further oxidation, is converted into /3-methyl-adipic acid —

CH3

CH8

•K

CH3

H2C CH2

H2C

CH2

H2C CO

H2C COOH

H2C

COOH

Xc/

COOH

C3H7

Menthone. Hydroxymenthylic acid. j8-Methyl-adipic acid.

1 Arth, Ann. Chim. PJtys,, 1886, VI., 7, 433; Beckmann and Mehrlauder,
Annalen, 1896, 289, 367.

THE MONO-CYCLIC TERPENES 63

These substances could be formed only if the isopropyl and
methyl radicals were in the para-position to each other ; for if
we take them in any other position, as shown below, the
resulting products are not the same —

CH3 CH3 CH3

CH CH CH

H2C CH.C3H7 H2C CO.C3H7 H2C COOH

HC CO H2C

2C CO H2C COOH H2C COOH

CH2

X CH2

Ketoacid. a-Methyl-adipic acid.

Again, the action of phosphorus pentachloride on menthone
gives a dichloro-tetrahydro-cymene,1 which, by successive treat-
ment with bromine and quinoline, produces a chlorocymene 2
of the constitution —

CH3

/\
HC CH

II I
HC C . Cl

10. The Syntheses and Constitutions of Menthol and Menthene.

Menthol is the alcohol corresponding to menthone, from
which it can be prepared by reduction. Since we have
established that menthone is (I.), it is obvious that menthol
must be (II.).

1 Berkenheim, Ber., 1892, 25, 694.

2 Jttnger and Klages, Ber., 1896, 29, 314.

64 RECENT ADVANCES IN ORGANIC CHEMISTRY
CH3 CH3

CH CH

•*--*-^-v'

HoC

H2C CH2 H2C CH2

I > I I

CO H2C CHOH

/ \ /

C3H.j

17 C3H7

(I.) (II.)

Now, when we dehydrate menthol, a hydrocarbon
^-menthene, is formed. This might be either (A.) or (B.), since
we can suppose that water is removed in either of two ways —
CH3

H2C

CH2

' \H

H2C

(JH

C CH

C3H,

17 C3H7

(A.) (B.)

The decision between the two formulae can be made by the
aid of the evidence of the oxidation products of menthene.1
When the menthene obtained from menthol is oxidized with
potassium permanganate solution, the first product is a glycol,
which, according to formula (A.), would have the constitution —

CH3

H2C CH2

I I

H2C CH— OH

VOH

C3H7

1 Wagner, Per., 1894, 27, 1639.

THE MONO-CYCLIC TERPENES 65

Further oxidation yields a ketone-alcohol, then hydroxy-
menthylic acid, and finally /3-methyl-adipic acid —

I I ~

H2C CO

0— OH

Ketone alcohol.

H2C COOH

C3H,

Hydroxymenthylic acid.

CH3

in K

HC/ XCH

.n2vy V>X12

H2C COOH

\
COOH

£-Methyl-adipic acid.

This is in agreement with the experimental results ; but if,
on the other hand, we start from the second possible formula
for menthene, the oxidation products would not be those found
in practice, but would be the compounds shown below—

CH3

H/ NCH

JlgO b±l

I II
H2C

C3H7

Thus the constitution of menthene must be —

CH3

He* r<tr
2v^ v_/.ti2

1 JH

CH8

CH3

/ \

CHOH

H2C COOH
y i

CHOH

H2C COOH

C3H,

H2C

66 RECENT ADVANCES IN ORGANIC CHEMISTRY

This has been confirmed by Wallaeh's recent synthesis of
menthene,1 in which he chooses as his starting-point 1, 4-methyl-
cyclohexanone (I.). This he condenses with a-bromo-isobutyric
ester by means of zinc, forming (II.) ; and then, by hydrolysis
and heating, causes the acid to lose carbon dioxide and become
converted into an alcohol (III.), which, on boiling with sul-
phuric acid, loses water and yields menthene.

H2C

CH3 CH

CH CH

' \ / N

CH2 H2C

0-tlq HoO

(CH3)2C OH

COOEt
(I.) (II.)

CH3 CH3

CH CH

H2C CH2 H2C CH2

H2C CH2 H2C CH

\ / \ *

C C

(CH3)2CH OH CH

CH3 CH£
(HI.) (IV.)

Menthene.

1 Wallaoh, Ser., 1906, 39, 2504.

THE MONO-CYCLIC TERPENES 67

11. The Constitution of Pulegone.

The last compound of the menthone group with which we
need deal is the unsaturated ketone pulegone.

If a ketone contains a double bond in the a]3-position to the
carbonyl group, hydroxylamine may react with it in two ways,
forming an oxime in the one case, and in the other attaching
itself to the double bond to give a hydroxylamine derivative.
For instance, in the case of mesityl oxide, we may have either
mesityl oxime or diacetone-hydroxylamine produced —

CH3.C:NOH CH3.CO

CH CH2

II I

C C.NH.OH

CH3 CH3 CH3 <JH3

Mesityl oxime. Diacetone-hydroxylaraine.

Now, since pulegone shows a similar behaviour, forming
either an oxime or a hydroxylamine derivative, the presump-
tion is that it also is a ketone with an unsaturated group in
the a/3-position to the carbonyl radical.

Again, pulegone on reduction is converted into menthone
so that it must contain the skeleton —

C C
C CO

And since we have found that it has the properties of an
unsaturated ketone it can have only three possible formulae

68 RECENT ADVANCES IN ORGANIC CHEMISTRY
CH3 CH3 CH3

CH C CH

H2C CH2 H2C CH H2C CH2

H2C CO H2C CO HC CO

C CH CH

CH3 CH3 CH3 CH3 CH3 CH3

(A.) (B.) (C.)

The evidence which enables us to choose between these
three has been supplied by Wallach,1 who has shown that
when pulegone is heated under pressure with water or anhy-
drous formic acid it undergoes decomposition into acetone and
methyl-cyclohexanone. Since this reaction can be explained
by Formula A alone, it is obvious that pulegone must have
that constitution. The break-down may be formulated in the
way indicated below —

CH3

H2C CH2

Yields methyl-cyclohexanone
H2C CO

C H2

C O Yields acetone

12. The Constitutions of the Phellandrenes.

The last hydrocarbon of the monocyclic class with which we
need deal is phellandrene, and it we must dismiss as briefly as
possible.

1 Wallach, Annalen, 1896, 289, 337.

THE MONO-CYCLIC TERPENES 69

Until a very short time ago, phellandrene was supposed to
be a simple substance, but in 1903 Semmler,1 from a study of
its oxidation products, was able to show that it must be a
mixture of two hydrocarbons, to which he gave the names
"normal phellandrene" and " pseudo-phellandrene" They are
also referred to as the a- and /3-forms of phellandrene in some
papers. Both have recently been synthesized, the normal form
by Harries and Johnson,2 and the pseudo-form by Kondakow
and Schindelmeister.3

The synthesis of the normal form is begun with the
substance A6-menthenone-2, but it may be well to show how
the constitution of this body is proved before we proceed
to the actual steps taken by Harries and Johnson. When car-
vone is treated with hydrobromic acid it forms a hydrobromide,
which, on reduction with zinc dust and methyl alcohol, gives the
required menthenone. As can be seen from the formulae below,
no doubt as to the constitution of the compound is possible.

CH3 CH2 CH3 CH3 CH3 CH<

Carvone. Carvone hydrobromide. A6-Menthenone (2).

When this menthenone (I.) is treated with phosphorus
pentachloride, its enolic form gives the substance (II.), which,
on reduction with zinc dust and methyl alcohol, gives
a-phellandrene —

1 Semmler, Ber., 1903, 36, 1749.

2 Harries and Johnson, Ber., 1905, 38, 1832.

3 Kondakow and Schindelmeister, /. pr. Ch., 1905, II. 72, 193; 1907, 76,
141.

70 RECENT ADVANCES IN ORGANIC CHEMISTRY
(I.) (II.) (III.)

i i

HC C . OH

i ii

HC C . Cl

^ I II

HC CH

^ \ ii

1 II
H2C CH

H2C CH

> II
H2C CH

\ /

/ \

<!„

/ \

/'ITT riTJ

V^JLls v^Xls

/ \
CH3 CH3

CHs^CH,

Menthenone. a-Phellandrene.

Enolic form.

The synthesis of the isomeric (3-phellandrene was actually
carried out by starting from tertiary menthene, but for the sake
of clearness we may go back in this case also to carvone. On
reduction, carvone (I.) gives dihydrocarveol and then carvo-
menthol (II.). This last substance, by loss of water, can be
converted into tertiary menthene (III.). By the action of
bromine upon this compound a dibromide (IV.) is formed,
which, with alcoholic potash, gives a hydrocarbon having all
the chemical properties of /3-phellandrene.

Carvone. Carvomenthol. Menthene.

C CH C

S \ /\ /\

HC CO H2C CH.OH H2C CH

HoO 0-tLo .UoG OHo JtioO

CH3 CH2 CH3 CH3 CH3 CH8

(I.) (II.) (m.)

THE MONO-CYCLIC TERPENES 7i

Dibromide. 0-Phellandrene.

Oil 3 CH2

I II

0— Br C

H2C CH.Br H2C OH

II II!

H2C CH2 H2C CH

i i

CH CH

CH3 CH8 CH8 CH3

(IV.) (V.)

With this substance we may conclude our review of the
monocyclic terpenes and turn in the next chapter to those
compounds which contain two rings of carbon atoms united
together.

CHAPTER IV

THE DICYCLIC TERPENES

A. — THE CAMPHENE GROUP
1. Syntheses of Camphoric Add

IN the series of dicyclic terpenes which we are about to describe
there are three important classes of substances. One group is
derived from the hydrocarbon camphene, another from pinene,
and a third from fenchene. Of these by far the most important
is the camphene group, with which we shall deal first. The
central substance of this group is the compound camphor
CioHi60 ; but in order to prove the constitution of this body it
will be necessary to proceed step by step, and in the first place
to prove the constitution of camphoric acid, which is obtained
from camphor by oxidation.

Komppa l and, later, Perkin and Thorpe 2 have synthesized
camphoric acid. We may deal with both of these syntheses,
beginning with the method employed by Komppa.

In this synthesis, the starting materials are oxalic ester and
/3/3-dimethyl-glutaric ester. These are condensed together with
sodium ethylate in the usual way, producing diketoapocamphoric
ester —

COOEt H . CH . COOEt CO CH— COOEt

| - 2EtOH

CHo.C.CH

CHo— C— CH

COOEt H . CH . COOEt CO CH— COOEt

Diketoapocamphoric ester.

This was then methylated by means of sodium and methyl
iodide, giving diketocamphoric ester —

1 Komppa, Ber., 1903, 36, 4332; Annalen, 1909, 368, 120; 370, 209; com-
l»are Blano and Thorpe, Trans. Chem. Soc., 1910, 97, 836.

2 Perkin and Thorpe, Trans. Chem. Soc., 1906, 89, 795.

THE DICYCLIC TERPENES 73

CO— — CH— COOEt

c — CH3

CO C COOEt

CH3

It is obvious that, since the formula is symmetrical, it
makes no difference which hydrogen atom is replaced by the
methyl group ; the end-product in each case is the same.

This diketo-ester was dissolved in sodium carbonate solu-
tion and then treated with sodium amalgam in a stream of
carbon dioxide ; by this means the two carbonyl groups were
reduced, and dihydroxycamphoric acid was formed, the ester
being hydrolyzed by the alkaline solution.

CH(OH) CH COOH

CH3— C— CH3
CH(OH) C COOH

Dihydroxycamphoric acid.

On boiling this substance with hydriodic acid in presence of
red phosphorus, it is converted into dehydrocamphoric acid,
which may have either of the constitutions shown below—

CH-.— Q COOH CH CH— COOH

| | |

CH3 — C — CH3 CH3 — C — CH3

CH2- -C COOH CH C COOH

CH3 CH3

Dehydrocamphoric acid.

The constitution of this acid is of no importance, however,
as the next two steps in the synthesis will yield the same final
product from either of the two acids formulated above. The
dehydrocamphoric acid is heated with hydrobromic acid in
acetic acid solution to 125° C., whereby it is converted into a
bromo-acid, which is then reduced with zinc dust and acetic

74 RECENT ADVANCES IN ORGANIC CHEMISTRY

acid to a substance which is identical with ordinary racemic
camphoric acid —

Br.CH

CH— COOH

CH2

G-H-3 — 0 —

- COOH

Bromo-acid.

CH2 - C - COOH

CHs

Racemic camphoric acid.

It will be seen at once that the exact constitution of the
dehydrocamphoric acid is of no great importance, as the posi-
tion of the bromine atom in the bromo-acid does not affect the
constitution of the final camphoric acid.

The synthesis of Perkin and Thorpe starts from tri-
methyl-1, 2, 2-bromo-l-cyclopentane carboxylic ester, which is
shaken with a mixture of potassium cyanide and hydrocyanic
acid solutions. The resulting substance is heated and then
boiled with acetic anhydride, whereby racemic camphoric
anhydride is formed.

KCN

Trimethyl-bromo-cyclopentane carboxylic ester. Camphoric acid.

One peculiarity of camphoric acid may be pointed out
here. An examination of the formula shows that camphoric
acid has two asymmetric carbon atoms in its ring — these are
distinguished by asterisks in the following formula : —

CHS

CH— COOH

C —

COOH

CH,

THE DICYCLIC TERPENES 75

Now, when we attempt to racemize dextro-camphoric acid by
any of the usual methods, it is found that instead of producing
an equimolecular mixture of dextro- and laivo-camphoric acids,
we obtain merely a mixture of dextro-camphoric acid with
a new substance, laevo-'iso-camphoric acid. From this be-
haviour of camphoric acid it is deduced that instead of both
asymmetric carbon atoms in the dextro-acid being inverted
(which would give us the mirror-image Isevo-camphoric) only
one is altered ; so that half the molecule remains as it was.
The change from ^-camphoric to ^-isocamphoric would be
represented thus —

CH3 CH3

CH2 C CH2 C

(CH3)2:C

COOH

(CH3)2:C

COOH
COOH

CH2 -- C CH2 ----- -C

COOH H

d-Camphoric acid. Z-Isoeamphoric acid.

2. The Synthesis of Camphor.

From synthetic camphoric acid we can obtain camphor
itself by the following method. When camphoric anhydride
is treated with sodium amalgam it is reduced to campholide,1
the reaction being analogous to the production of phthalide
from phthalic acid.

CH3 CH3

Camphoric anhydride. Campholide.

1 Haller, Bull soc. cliim., 1896, [Hi.] 15, 7, 984 ; Forster, Trans. Chem. Soc.t
1896, 69, 36.

76 RECENT ADVANCES IN ORGANIC CHEMISTRY

Campholide, on treatment with potassium cyanide, produces
a nitrile-salt, which, on hydrolysis, gives homocamphoric acid l ;
this action is exactly like that which produces homophthalic
acid from phthalide —

CH2 CH CH2

CH3— C— CH3 "o
OHo C CO

CH3

Nitrile-salt.

Campholide.

CH3 — C — CH3
CH2 -- C --- COOH

CH3

Homocamphoric acid.

From this homocamphoric acid it is easy to produce camphor
itself by distilling the lead or calcium salt of the acid.2

CH2 CH— CH2 . COO

I I - CaC03

CH3— C— CH3 Ca >

CH, C COO

CH-

CHc

This synthesis confirms the camphor formula which was
put forward in 1893 by Bredt.3

3. Borneol, Camphene, and Camphane.

When camphor is reduced by means of sodium and alcohol 4
it yields a secondary alcohol, borneol, which has the formula —

1 Haller and Blanc, C. R., 1900, 130, 376.

2 Haller, C. R., 1896, 122, 446; Bredt and Rosenberg, Annalen, 1896, 289, 5-

3 Bredt, Ber., 1893, 26, 3047.

4 Jackson and Menckc, Am. Chem. J., 1883, 5, 270; Wallach, Annalen, 1885,
230, 225.

THE Dl CYCLIC TERPENES 77

CH.OH

OHg

Borneol.

This alcohol occurs in dextro- and laevo-forms, either of
which may be obtained at will by reducing the corresponding
dextro- or laevo-camphor. Borneol is not the only product of
this reaction, however, as at the same time a small quantity of
an isomeric isoborneol 1 is produced, whose constitution is not
yet definitely proved.

The hydroxyl radical in borneol can be replaced by a
halogen atom in the usual way,* and if the bornyl iodide thus
formed be reduced by means of zinc dust, acetic and hydriodic
acids,2 a hydrocarbon camphane, is produced, which is the root-
substance of the camphor series. It has the formula —

CHa

Camphane.

On the other hand, when bornyl chloride or bromide is
heated with alcoholic potash it is converted into an unsaturated
substance by the loss of a molecule of a halogen acid.3 The
constitution of this new hydrocarbon, camphene, CioHie, is as
yet undetermined. The simplest possible constitution would
be the one shown below ; but this has been proved to belong to
bornylene,4 which is produced by the long-continued action of

1 Montgolfier, C. R., 1879, 89, 101 ; Haller, C. R., 1887, 105, 227.

* In practice, however, bornyl iodide is usually prepared by the action of
hydriodic acid on pinene, as the yields from borneol are very poor.

2 Aschan, Ber., 1900, 33, 1006.

3 Riban, Ann. Ghim. Phys., 1875, V. 6, 353.

4 Wagner and Brjckner, £er., 1900, 33, 2, 21.

78 RECENT ADVANCES IN ORGANIC CHEMISTRY

alcoholic potash upon bornyl iodide. This substance, bornylene,
on oxidation yields camphoric acid, which camphene does not do.

CH3 — C — CH3

^1TT f1 (

^-i-i-2

JHOH
^H

CH3
Borneol.
~TT flH C

CH3-C— CH3

CH,

-CH

CFT.I

CH3

Bornyl iodide.

•CH COOH

COOH

Bornylene.

CH3

Camphoric acid.

The oxidation products of camphene are much more compli-
cated, and will require a section to themselves. Before dealing
with them, however, we must take up the question of the
oxidation of camphor itself.

4. The, Decomposition Products of Camphoi\

The most vulnerable point in the camphor molecule is the
carbonyl group and the adjacent methylene radical. The ring at
this point is so easily attacked that it may be broken by a simple
hydrolytic reaction. When camphor is heated with sodium
and xylene to a temperature of 280° C., the ring opens ; and
when the reaction mixture is poured into water, the sodium
salt of campholic acid l is formed.

CHa— C— CH3

-GH CHc

CH3 — C — CHs

CHo

C COOH

CHc

Camphor. Campholic acid.

1 Malin, Annakn, 1868, 145, 201 ; Kachler, ibid., 1872, 162, 259.

THE DICYCLIC TERPENES

The same acid has been obtained by Haller and Blanc 1 from
campholide, a method of synthesis which establishes the consti-
tution of the substance beyond doubt.

CO CH2 CH2Br CH3

/ \ +4H / \ HBr / 2H /

H14 0 — ^ C8H14 0 — > C8H14 — > C8H14

COOH

Camphoric anhydride. Campholide.

Bromocampholic Campholic
acid. acid.

Now, when campholic acid is oxidized with nitric acid, the
newly formed methyl group is oxidized to carboxyl, and
camphoric acid is formed.

CH,

-CIL

CHo

--COOH

CH2 CH COOH

CH3-C— CH3
CH2 C COOH

CH3 OH3

Campholic acid. Camphoric acid.

Further action of nitric acid upon the latter substance
gives rise to camphanic acid, which is oxidized in its turn to
camphoronic acid —

^-iA2 — — \j±i — — \j\j\jj^. V^X12 V

j uuun

^"0

CH3— C— CH3 >

CH3— C— CH3 |

3H2 (

3 COOH <

M. <

CHc

Camphoric acid.

COOH

CH3— C— CH

CH3

Camphanic acid.

COOH

-COOH

CH3

Camphoronic acid.

1 Haller and Blanc, G. R., 1900, 130, 376.

8o RECENT ADVANCES IN ORGANIC CHEMISTRY

The constitution of cainphanic acid 1 is proved by the fact
that it can be obtained from bromocamphoric anhydride by
boiling with water —

CH2 C CO

IN, \

CH3— C— CH3 0
CH2 C CO

CH3

Bromocamphoric anhydride.

C COOH

CH3

Camphanic acid.

The constitution of camphoronic acid was established by the
synthesis of Perkin and Thorpe.2 These authors first prepared
/3-hydroxy-trimethyl-glutaric ester by the action of zinc upon
a mixture of acetoacetic ester and a-bromo-isobutyric ester, or
upon a mixture of dimethyl-acetoacetic ester and monobrom-
acetic ester —

(CH3)2C . Br CO CH2

COOE CH3 COOlN*(CH3)2C C(OH)— CH2

/COOE CH3 COOE

COOE CH3 COOE j8-Hydroxytrimethylglutaric ester.

By replacing the hydroxyl group first with chlorine and then
by cyanogen they obtained the nitrile-ester of camphoronic acid,
from which the acid itself was produced by hydrolysis —

(CH3)2C C(CH3)— CH2 (CH3)2

COOE CN COOE

Camphoronic nitrile.

C(CH3)— CH2

COOH COOH COOH

Camphoronic acid.

When camphoronic acid is heated to above 135° C., it loses
water and is converted into anhydrocamphoronic acid, C9Hi205.

1 Key her, "Dissertation," Leipzig, 1891; Bredt, Ber., 1894, 21, 2097;
Lapworth and Lenton, Trans. Chem. Soc., 1902, 81, 17.

2 Perkin and Thorpe, Trans. Chem. Soc,, 1897, 71, 1169,

THE DICYCLIC TERPENES 81

By brominating the chloride of this acid, two isomeric bromo-
anhydrocamphoronic chlorides are produced, one of which, when
boiled with water, gives the lactone of an unstable hydroxy-
camphoronic acid (camphoranic acid), while the other yields
stable hydroxycamphoronic acid. Camphoranic acid, when
fused with potash, breaks down into oxalic and trimethyl-
succinic acids.1 These changes may be expressed thus —

COOH COOH COOH COOH

CH3— C - CH2 CH3— C- -CH
CHo -- > CH3— C— CH3

C—

COOH CO O

Camphoronic acid. Campboranic acid.

COOH COOH
CH3— CH COOH
^ CH3 — C — CH3

COOH

Trimethylsuccinic and oxalic acids.

5. Camphoic and Apocamphoric Acids.

Camphene contains a double bond, by means of which it
unites with halogen acids. When it is oxidized by means of
dilute potassium permanganate, the usual addition of hydroxyl
groups at each end of the double bond occurs, with the
formation of camphene glycol,2 CioHi6(OH)2; but if for the
permanganate we substitute nitric acid, the first product
isolated is camphoic acid.3 Just as malonic acid on dry
distillation loses carbon dioxide and is converted into acetic
acid, so camphoic acid loses carbon dioxide and yields
apocamphoric acid. From this we may deduce that cam-
phoic acid is a tribasic acid of the constitution shown below —

1 Bredt, Annalen, 1898, 299, 150.

2 Wagner, Ber., 1890, 23, 2311.

3 Marsh and Gardner, Trans., Chem. Soc., 1891, 59, 61; 1896, 69, 74.

82 RECENT ADVANCES IN ORGANIC CHEMISTRY
CH2 CH COOH CH2 CH COOH

CH3— C— CH3

CH3— C— CH

OH2- -C- -COOH CH2- -CH COOH

COOH

Camphoic acid. Apocamphoric acid.

The constitution of apocamplioric acid is established by
the fact that it can be prepared by the reduction, under
suitable conditions, of diketoapocamphoric acid with which we
have already dealt.

We must now leave the camphor group and turn to the
isomeric substances in the fenchene and pinene series.

B. — FENCHONE AND ITS DERIVATIVES.
1. The Constitution of Fenchene.

Fenchone is a ketonic compound, isomeric with camphor,
and resembling it in many respects. When dextro-fenchone is
reduced it yields D-Z-fenchyl alcohol1; the name indicates
that though derived from a dextro-ketone the substance is
actually laevo-rotatory. When this is treated at a low temperature
with phosphorus pentachloride it gives laevo-rotatory D-Z-fenchyl
chloride; which, by the action of aniline, loses hydrochloric
acid and is converted into D-Z-fenchene, just as bornyl chloride
is changed into carnphene. If the reaction mixture during
the formation of fenchyl chloride be allowed to grow warm,
the resulting substances are not D-Z-fenchyl chloride and
D-Z-fenchene, but D-tZ-fenchyl chloride and D-tZ-fenchene.

The constitution of D-Z-fenchene has been dealt with in the
following way.2 When it is oxidized with potassium permanga-
nate it is converted into a hydroxy-acid, D-Z-hydroxy-fenchenic
acid, which has the composition Ci0Hi603. This body, when
treated with lead peroxide and sulphuric acid, loses carbon
dioxide and two atoms of hydrogen, being converted into
D-^-fenchocamphorone, CgH^O. By nitric acid this last com-
pound is broken down to apocamphoric acid. This production

1 Wallach, Annalen, 1891, 263, 143.

2 Wallach, Annalen, 1898, 300, 294; 1901, 315, 283.

THE DICYCLIC TERPENES

of apocamphoric acid from fenchene shows that in fenchene
itself one of the carbon atoms must be attached to the nucleus
at a point different from that at which the methyl group in
camphor is placed as otherwise we should find camphoric acid
produced in the end instead of its next lower homologue, apo-
camphoric acid. The only way in which we can satisfy this
requirement is shown in the formulae below —

CH2 CH C : CH2 C]

CH3— C— CH3

CH C(OH). COOH

CH3-C— CI
— CH-

CH2 CH CH2 CH2 —

D-Z-fenchene. Hydroxyfenchenic acid.

CH2

-CH CO

CH,

-CH . COOH

CH3— C— CH3 | — -
CH-

CH3— C— CH3

I I

CH2 CH.COOH

Apocamphoric acid.

Fenchocamphorone.
D-Z-fenchene, therefore, has the constitution expressed by —
CH2— -CH— -C : CH2

CH3 — C — CH3
CH2 CH CH2

2. The Constitution of Fenchone and Fenchyl Alcohol.

Claisen l has shown that when ketones containing a methyl-
ene group next the carbonyl radical are treated with sodium
and amyl formate they are converted into oxymethylene
derivatives —

E— CH2
R— CO

R_C : CH . OH

E— CO

Camphor when dealt with in this manner forms oxymethylene
1 Claisen, Annalen, 1894, 281, 394.

84 RECENT ADVANCES IN ORGANIC CHEMISTRY

camphor ; but when we apply the same reaction to fenchone no
such result is obtained. From this we conclude that fenchone
contains no methylene group next the carbonyl radical.

We have just proved the formula of D-Z-fenchene, and from
it we can deduce that of the corresponding saturated compound,
fenchane —

}H C : CH2 CH

CH3— C-CH3
-CH

CH . CH3

D-Z-fenchene.

Now, fenchone is derived from fenchane by replacing a
methylene group by a carbonyl radical, and, in accordance with
what we have just shown, the carbonyl group so produced must
have no methylene group adjacent to it. There is only one
formula which fulfils these conditions, so that the constitution
of fenchone must be expressed by —

CH2

CH . CH3

CH3 — C-^CH3

—in—

Fenchone.

Since fenchyl alcohol is obtained by the reduction of
fenchone, its constitution must be that which is shown below —

CH2 CH CH.CHg CH2

CH3 — C — CH3

CH2 CH —

Fenchone.

-CO

-CH—
-C~

CH.CH3

CH,

-CH CH . OH

Fenchyl alcohol.

Another formula for fenchone has been suggested by
Semmler1 and supported by Bouveault and Levallois2—

1 Semmler, Ch. Zig., 1905, 29, 1313.

2 Bouveault and Levallois, C. r., 1908, 146, 180.

THE DICYCLIC TERPENES

CH,

OH,

-CH-

CH2

-0—

OH,

•C(CH3)2

-CO

This formula explains certain reactions which Wallach's
does not make clear. The matter is still under discussion.

C.— PINENE.

1. The Constitution of Pinene.

Pinene is a hydrocarbon isomeric with camphene and
fenchene. It was found by Sobrero 1 that when this substance
was allowed to stand in sunlight in contact with water and
air it was, after several months, converted into a compound
sobrerol, CioHj6(OH)2, which, on boiling with dilute acids, was
changed, by the loss of one molecule of water, into pinol,
CioHieO. Pinol was found, on further investigation, to be an
internal ether of the same type as cineol. Wallach 2 has shown
that pinol may also be obtained by the action of sodium ethylate
on terpineol dibromide.

When pinol or sobrerol is treated with a one per cent, solution
of potassium permanganate the product is a dihydric alcohol,8
pinol-glycol, CioHi60(OH)2. On further oxidation, a tetra-
hydric alcohol,4 sobrerythrite, CioH^OH)*, is formed, which in
turn is oxidized to terpenylic acid. Therefore we should find
in pinene, pinol, and pinol-glycol, the same chain of carbon
atoms which we know exists in terpenylic acid—

COOH

1 Sobrero, Annalen, 1851, 80, 106.

2 Wallach, Annalen, 1890, 259, 309.

8 Wagner and Slawinski, Ber, 1894, 27, 1644.

4 Wagner and Ginsberg, Per., 1894, 27, 1648; 1896, 29, 1195.

86 RECENT ADVANCES IN ORGANIC CHEMISTRY

In other words, the pinol skeleton must contain the
grouping—

CH2 CH CH2

CH3 — C — CH3

CH =

= C

Into this scheme we have now to fit a hydrogen atom and
the group —

CH*— C :

and, as can at once be seen, there are two possible ways of
doing this —

CH2 CH

CH3— C-CH3

CH2

-C-

CH3

Pinol T.

CH CH,

CH3— C— CH3

\
CH2 C== =0

CH3

Pinol II.

On these two assumptions sobrerol, which is obtained from
pinol by the addition of water, would have either of the
formulae —

US.1%

Cl
"!TT

?a — C — CH3
OH

o r

H.OH (

CH3 — C — CH3

^H C (

LOH

CH3

Sobrerol l.a.

CH3
Sobrerol Il.a.

Now, sobrerol, on oxidation with a one per cent, solution of
potassium permanganate, gives a tetrahydric alcohol, sobrery-

THE DICYCLIC TERPENES

thrite. This can only be explained by using the formula (I.a),
for (Il.a) would produce a hydroxy-ketone —

CHS

-CH-

-CH,

— C — CH3

CH2 CH-

-CHc

3. OH

CH3 — C — CH3

CH2 CO COOH

CH3

Hydroxyketone.

Sobrerol, therefore, has the formula (La) and pinol the
formula (I.).

From this we may conclude that the formula of pinene
itself is —

CH2 CH CHS

H3
Sobrerol Il.a.

CH2

Pinene.

In virtue of the double bond in its molecule, pinene is
capable of uniting with hydrochloric acid or nitrosyl chloride.
Pinene hydrochloride resembles camphor in appearance and
smell, and is used commercially under the name of " artificial
camphor." Pinene nitroso-chloride,1 on standing in presence of
hydrochloric acid, is converted into hydrochlorocarvoxime by
the wandering of a chlorine atom and the rupture of the pinene
tetramethylene ring.

0x1

C H3 — C — C H3

C-

Pinene nitrosochloride.

N.OH CH3

Hydrochlorocarvoxirae.

Baeyer, Ber.t 1896, 29; 20.

88 RECENT ADVANCES IN ORGANIC CHEMISTRY

Pinene itself is converted into terpineol by hydration with
dilute acids —

J OJtl Oll9

CH3— C— (
OH

^iA2 - V^/XJ. -

CH 3 — 0 — v^H3

;AJL2

ITT r f

CH3

Pinene.

CH;

Terpineol.

2. Pinonic and Pinic Acids

When pinene is oxidized with potassium permanganate,
the first product is a ketonic acid l which, according to the
conditions of the experiment, can be obtained either as a single
substance or as a mixture of two isomers. When the single
substance is produced it is found to have the composition
CioHi603, and has been named a-pinonic acid. It contains the
group CH3 — CO — , for, on treatment with bromine and potash,
it loses a methyl group, takes up hydroxyl, and is converted
into pinic acid, CgHuCV-

KOH and Br

C8Hi302— CO— CH3 - > C8Hi302— COOH + CHBr8

Pinonic acid. Pinic acid.

These changes are expressed in the following formulae : —
CH2 CH CH2 CH2 CH CH2

CH3— C— CH3
==0 — CH

CH3— C— CH3
COOH CO -CH

CH3

Pinene.

CH8

Pinonic acid.

CH2 CH CH2

CH3 — C — CH3

COOH HOOC CH

Pinic acid.

1 Baeyer, Ber., 1896, 29, 3.

THE DICYCLIC TERPENES

Now, on hydrolysis with fifty per cent, sulphuric acid,
pinonic acid gives a keto-lactone,1 Ci0H1602, which proves to be
identical with that obtained in the oxidation of terpineol. A
similar hydrolysis converts pinene into terpineol, so that we
may draw up the following scheme to show the relations
between the four substances : —
CH2 CH Cttj

CH3— C-
CH

CH3

^ hydrolysis
CH2 CH CH2

CH3-C-CH,

Pinene.

CH3 oxidation

CH CH2

CH3— C— CH3 I Pinonic acid.
OH CO CH

Terpineol.

oxidai
OH

oxidation

Ketolactone.

CH2

D. — BORNYLENE AND THE THUJENES
In this section we may deal very briefly with the two

substances, bornylene and thujene.

Bornylene is obtained from bornyl iodide by the action of

alcoholic potash. On oxidation it gives camphoric acid. From

this we can deduce that its formula must be that which is

shown below —

CH2 CH CH2

OHq 0 OHq

CH,

CH.I

CH3

Bornyl iodide.

CH5

-/-LJ-2 ~ \JJ-1 v

CH3— C— CH3
^TT r r

/a j.
JH

CH3

Bornylene.

CH COOH

CH3— C— CH3
CH9 C COOH

CH3

Camphoric acid.

1 Baeyer, Ber., 1896, 29, 3.

90 RECENT ADVANCES IN ORGANIC CHEMISTRY

Thujone is a ketone isomeric with camphor. Its consti-
tution has not yet been established by a synthesis, but at the
present time it appears almost certain that it has the formula
ascribed to it by Semmler,1 in which it is represented as a
combination of a six-membered with a three-membered ring —

CH3 CH3

¥ i

C IT2 — C

CH — CH— CO
I
CH3

When thujone is reduced it yields thujyl alcohol ; and from
this we can produce the thujyl derivative of xanthogenic acid
(sulphothio-carbonic acid). When this substance is distilled it
breaks down into carbon oxysulphide, methyl mercaptan, and a
hydrocarbon, thujene —

C10H170 . OS . SCH3 = COS + CH3SH -f C10H16

Tschugaeff 2 has shown that thujene thus obtained is a mixture
of two hydrocarbons, to which he attributes the formulae—

03117 03x17

CH2— C— CH2 OH2-C --- CH

I I \/\

CH — C=CH CH— CH — CH

CH3 CH3

a-Thujene. £-Thujene.

1 Semmler, Ber., 1900, 33, 275, 2459.

2 Tschugaeff, Ber.t 1901, 34, 2279; 1904, 37, 1481.

CHAPTEE V

THE OLEFINIC TERPENES

A. — INTRODUCTION

WE have now described all the important cyclic terpenes, and
in pursuance of the plan laid down in the first section dealing
with these bodies, we must next examine the olefinic substances
which are often included in the terpene group. It might have
been more logical to have dealt with the open-chain compounds
first, and the cyclic ones later, but as we should in that case
have had to assume the constitution of certain cyclic terpenes
which are closely connected with the olefinic ones, the present
method of arrangement is more convenient.

Those unsaturated open-chain substances which are found in
ethereal oils, and which, in many cases, can be transformed into
cyclic terpenes, are termed olefinic terpenes, or terpenogens.
They occur as hydrocarbons, aldehydes, or alcohols, and are
derived from hydrocarbons of the formula C5H8. In many
cases the odour of ethereal oils is very largely due to the olefinic
terpenes contained in them.

The chemical importance of the olefinic terpenes lies in the
fact that from them we can build up some of the more
complicated terpene derivatives by means of very simple
reactions ; but they are of interest also from the commercial
point of view as forming the basis of many natural and
artificial perfumes.

B. — ISOPRENE.

Isoprene is the simplest of all the olefinic terpenes; it
contains two double bonds, and has the composition C5H8. Its
synthesis has been carried out by Euler,1 and also by Ipatjew,2

1 Euler, J. pr. Ch., II. 57, 132.

2 Ipatjew, ibid., 55, 4.

92 RECENT ADVANCES IN ORGANIC CHEMISTRY

in the one case starting from methyl-pyrrolidine, and in the
other from dimethyl-allene. In the first case, the inethyl-
pyrrolidine (I.) is allowed to interact with methyl iodide with
the formation of dimethyl-methylpyrrolidinium iodide (II.).
This substance is then decomposed with potash, whereby the
ring is broken and fe-dimethyl-methylpyrrolidine (III.) is
produced. The addition of methyl iodide and decomposition of
the product (IV.) with potash gives trimethylamine and the
required isoprene (V.).

a.) (no cm.)

CH3— CH— CH2 CH8— CH— CH2 CH3— OH— CH

CH2 CH2 CH2 CH2 CH2 CH2

\/ \

CH3— N— I N(CH3)2

CH,

(IV.) (V.)
CH3— CH— CH CH0 - C CH

—

. CH2

N(CH3)3

+ N(CH3)3 + HI

The synthesis from dimethyl-allene is much simpler. Two
molecules of hydrobromic acid are added on, forming 2- methyl-2,
4-dibromobutane, from which hydrobromic acid is again split off
by means of alcoholic potash —

CH3 CH3 CH3

C : C : CH2 CBr . CH2 . CH2Br C . CH : CH2

/ / S

riTT r^TT f^TT

v^/jLi3 v-'-tis v-^-da

Dimethyl-allene. Methyl-dibromobutane. Isoprene.

Isoprene is produced by the dry distillation of indiarubber
and by the decomposition of turpentine oil at a dull red heat.
Concentrated hydrochloric acid converts it into a polymer which
has all the physical properties of indiarubber, and the same
change takes place on long standing or with traces of acids in

THE OLEFINIC TERPENES 93

sunlight. When heated to 300° C., isoprene is polymerized to a
di-isoprene, which seems to be identical with dipentene.1

i c

/\ /\

H2C OH H2C CH

II I I

H2C OH2 H2C CH2

\ \/

CH CH

0x12

In a somewhat similar manner isoprene might be supposed
to give a sesquiterpene in which three isoprene molecules would
coalesce to form a compound of the composition Ci5H24. In any
probable reaction of this type, it is worth noting, at least one
unsaturated chain will be left untouched and ready to react
with further molecules if the proper conditions are obtained ;
and it is doubtless to this side chain that we owe the more
complex polymer which resembles indiarubber.

C.— ClTKONELLAL.

We must now pass to the consideration of a substance rather
more complicated than isoprene — the compound citronellal,
which was discovered by Dodge 2 in citronella oil. Citronellal
is an aldehyde, for on reduction it gives the alcohol citronellol,
and on oxidation it forms citronellic acid. Since it is optically
active it must contain an asymmetric carbon atom.

Tiemann and Schmidt,3 oxidizing it in aqueous solution,
obtained as products acetone and |3-methyl-adipic acid, from
which they concluded very naturally that citronellal had the
constitution —

1 Tilden, Trans. Chem. Soc., 1884,45,410 ; Bouchardat, C. R., 1875, 80, 1446 ;
1878, 87, 654 ; 1879, 89, 361, 1117.

2 Dodge, Am. Chem. J., 1889, 11, 456.

8 Tiemann and Schmidt, Per., 1896, 29, 903 ; 1897, 30, 22, 33.

94 RECENT ADVANCES IN ORGANIC CHEMISTRY
(CH3)2C = CH . CH2 . CH2 . CH(CH3) . CH2 . CHO

(CH3)2CO + CH2 . CH2 . CH(CH3) . CH2 . COOH
COOH

The reason for placing the methyl group in this position will be
seen later when we deal with the production of pulegone from
this body.

This constitution, however, is not in agreement with the
work of Harries and Schauwecker,1 who approached the matter
from a slightly different standpoint. Instead of oxidizing
citronellal itself, they prepared its dimethyl-acetal and replaced
the aqueous solution of Schmidt and Tiemann by an acetone
one. Under these circumstances they found that the oxidation
product with potassium permanganate was the acetal of a
dihydroxy-dihydrocitronellal, which, on further oxidation with
chromic acid, could be converted into a keto-aldehyde. This
shows that the double bond must lie at the extreme end of the
chain, so that citronellal would have the constitution —

CH3 CH3

C . CH2 . CH2 . CH2 . CH . CH2 . CHO

CH2

On this view the dihydroxy-compound and the keto-aldehyde
would be —

CH3 CH3

C(OH) . CH2 . CH2 . CH2 . CH . CH2 . CHO

CH2OH

CH3 CH3

CO . CH2 . CH2 . CH2 . CH . CH2 . CHO

The results obtained by Tiemann and Schmidt would be
explained by supposing that under the influence of the aqueous
oxidizing agent the position of the double bond was changed

1 Harries and Schauwecker, Ber., 1901, 34, 1498, 2981.

THE OLEFINIC TERPENES 95

from the ultimate to the penultimate pair of carbon atoms
in the chain.

So far we have not proved the position of the methyl group,
but we shall now give some evidence bearing upon the point.
When citronellal is allowed to stand by itself for a considerable
time it is converted into the isomeric substance isopulegol.1
The same change is brought about more rapidly by heating
citronellal with acetic anhydride 2 to 180° C. The change may
be represented in the following manner : —

CH3 CH3

CH CH

H2C CH2 H2C CH2

II II

H2C CHO H2C CH.OH

CH2 CH

I I

r\ r\

CH3

Citronellal. Isopulegol.

The proof of the constitution of isopulegol depends upon its
conversion into pulegone. When it is oxidized it yields the
ketone isopulegone, which is converted into pulegone by the
wandering of a double bond.

CH3 CH3 CH3

CH CH

H2C CH2 H2C CH2 H2C CH2

H2C CH.OH H2C CO H2C CO

\ / \ / \ /

CH CH C

I I II

C C C

/ \ / \ : 1 / \

CH3 CH-2 CH3 OH 2 C-tL3 CJti3

Isopulegol. Isopulegone. Pnlegone.

1 Labbe, Bull.soc. chim., 1899, [ill] 21, 1023.

2 Tiemann and Schmidt, Ber., 1896, 29, 913; 30, 27

96 RECENT ADVANCES IN ORGANIC CHEMISTRY

From this it is evident that the methyl group in citronellal
must be in the position which we attributed to it as otherwise
the isopropylene group would not come into the 1, 4-position
with it in the pulegone formed from citronellal.

We may postpone the consideration of the alcohol citronellol
and of citronellic acid until later, as they are closely connected
with some members of the class of compounds with which we
are about to deal in the next section.

D. — THE CITEAL GROUP.
1. General.

The group of olefmic terpenes, of which citral is the most
important member, can all be derived from the unsaturated
ketone methyl-heptenone. It will perhaps be best, before
entering upon a detailed consideration of the group to give
a small table showing the relations between the different
members.

Methylheptenone

acid

Citral Rhodinic acid

/ \ / \ I

Geraniol Nerol Rhodinol Rhodinal

. *

Linalool

We must now proceed to trace out the various changes by
which the several substances are obtained.

2. Methyl-heptenone.

As can be seen from the foregoing table, the substance from
which all the other members of the citral group are built up is
the ketone methyl-heptenone. We have already encountered
this compound among the decomposition products of cineolic
acid, but in that place we did not deal with its constitution.

Methyl-heptenone has been synthesized in different ways

THE OLEFINIC TERPENES 97

by Barbier and Bouveault,1 Verley,2 Tiemann,3 Leser,4 and
Ipatjew.5 We need only give one synthesis here, and may
choose that of Barbier and Bouveault. In the first place,
2-methyl-2, 4-dibromobutane is condensed with the sodium
derivative of acetylacetone. This gives the unsaturated dike-
tone (II.), which can be broken down by strong alkali into
acetic acid and methyl-heptenone (III.).

(CH3)2C.Br (CH3)2C (CH3)2C

CH2 CH CH

CH2Br CH2 CH2

CH3 . CO CH3 . CO CH3 . CO . CH2
CH.Na CH

CH3.CO CH3.CO CH3.CO.OH

(I.) (II.) (HI.)

This establishes the constitution of the substance, but if
further proof were required it is to be found in the behaviour of
methyl-heptenone (A) on oxidation. The first product (B) is a
dihydroxy-ketone, which, on further oxidation, breaks down
into acetone and laevulmic acid (C).

C C— OH

CH3
CH > CH . OH

CH2 . CH2 . CO . CH3 CH2 . CH2 . CO

(A.) (B.)

1 Barbier and Bouveanlt, C. #.,1896, 122, 393.
8 Verley, Bull soc. eta., 1897, [iii.] 17, 180.

3 Tiemann, Ber., 1898, 31, 824.

4 Leser, Bull. soc. chim., 1897, [iii.] 17, 180.

5 Ipatjew, Ber., 1901, 34, 594.

98 RECENT ADVANCES IN ORGANIC CHEMISTRY
CH3

— » v: . ;: •

CHg

COOH

CH2.CH2.CO.CH3

(C.)

In itself, methyl-heptenone is of no great importance, and
we may confine ourselves to one of the reactions which it
undergoes. When shaken with seventy-five per cent, sulphuric
acid it loses a molecule of water and is converted into
dihydro-m-xylene —

HBC

CH CH

/ \ S \

CH3— C CH2 CH3— C CH2

I -H20 | |

CH2 > HC . CH2

0:C XC

CH3 CH3

Methylheptenone. Dihydro-w-xylene.

3. Geranic Acid.

Following upon their synthesis of methyl-heptenone,
Barbier and Bouveault 1 were enabled to synthesize geranic acid
by means of a simple series of reactions with which we must
now deal. By the action of zinc and iodo-acetic ester upon
methyl-heptenone they prepared a hydroxy-acid, which, on
boiling with acetic anhydride, broke down into geranic acid.

31 825 ' R" 1896' 122> 393 •' seealso Tiemann, Per., 1898,

THE OLEFINIC TERPENES 99

The formulae below indicate the course of the synthesis —

I ?H'

(CH3)2C : CH . CH2 . CH2 . CO

I Zinc and iodoacetic ester

CH3
(CH3)2C : CH . CH2 . CH2 . C . OZnl

Methyl-heptenone.

Intermediate product.

CH2 . COOEt

Water

CH3

)H3)2C : CH . CH2 . CH2 . C . OH Hydroxydihydrogeranic ester.

CH2. COOEt
Hydrolysis

(CH3)2C : CH . CH2 . CH2 . C . OH Hydroxydihydrogeranic acid.

CH2COOH

} Dehydration with acetic anhydride

CH3

(CH3)2C : CH . CH2 . CH2 . C : CH . COOH Geranic acid.

Like methyl-heptenone, geranic acid is of very little
importance in itself. The only reaction which specially con-
cerns us is its condensation to a-cyclogeranic acid,1 which, like
the corresponding condensation of methyl-heptenone, takes
place under the influence of seventy per cent, sulphuric acid.
In order to explain the geranic acid change, it is necessary to
assume the formation and decomposition of an intermediate
product which has not yet been isolated —

1 Tiemann and Semmler, Ber.t 1893, 26, 2726 ; Tiemann and Schmidt, ibid.,
1898, 31, 881; Tiemann and Tigges, ibid., 1900, .33, 3713; Barbier and
Bouveault, Bull. soc. chim., 1896, [iii.] 15, 1002,

TOO RECENT ADVANCES IN ORGANIC CHEMISTRY

CH3 CH

I II
H2C 0 .

CH2

CH3 CH3
VoH
oC CH2COOH

o-Cyclogeranic acid.

CH3 CH3

CH . COOH H2C CH . COOH

II ^ CH3 — -> | |

/ H2C C . CH3

H2C C \ //

\/\
CH2 OH

Geranic acid. Hypothetical

intermediate product

As the table shows, geranic acid gives rise to two series of
compounds ; on the one hand, by reduction, we may obtain
rhodinic acid and its derivatives, while on the other we may
produce the aldehyde citral, from which in turn several sub-
stances may be formed. In the first place, we may deal with
the smaller group, rhodinic acid and its allied compounds.

4. Rhodinic Acid, Rhodinol, and RhodinaL
When the ethyl ester of geranic acid is reduced by means
of sodium and amyl alcohol it is converted into inactive rhodinic
acid.1 The active, laevo-rotatory form of this acid has been
obtained from the active alcohol rhodinol. These two acids
are isomeric with citronellic acid, which is obtained by the
oxidation of the aldehyde citronellal, and it has been suggested
that citronellic acid is the dextro-form of rhodinic acid. On
the other hand, from the constitution of citronellal, we should
expect that citronellic acid obtained from it by oxidation would
have the formula (I.), while rhodinic acid from geranic acid
should have the formula (II.).

CH2 : C . CH2 . CH2 . CH2 . CH . CH2 . COOH

CH8 CH3

Citronellic acid.

(I.)

(CH8)aO : CH . CH2 . CH2 . CH . CH2 . COOH

CH3

Rhodinic acid.

(ii.)

1 Tiemann, Ber., 1898, 81, 2901.

THE OLEFINIC TERPENES 101

The literature of the subject is somewhat contradictory, and

it does not seem necessary to go into the question in detail here.

When the ester of rhodinic acid is reduced by means of

sodium and absolute alcohol it yields the corresponding alcohol 1

rhodinol —

(CH3)2C : CH . CH2 . CH2 . CH . CH2 . CH2OH

CH3

which is isomeric with citronellol. Here, again, the literature
is contradictory, and it seems impossible to decide whether the
two compounds are stereo-isomers or differ in structure.

Khodinal,2 the aldehyde corresponding to the alcohol
rhodinol, is obtained by distilling together calcium formate and
the calcium salt of rhodinic acid. Barbier and Bouveault regard
it as having the structure (I.), because of its conversion into
menthone. Citronellal, with which it is isomeric, when sub-
mitted to the action of acetic anhydride, is changed into
isopulegol, as we have already described. On the other hand,
rhodinal when treated in the same way yields menthone —

CH3 CH3

OH OH

H2C CHO H2C CO

\ \ /

CH CH

II I

C CH

CH3 CH3 CH3 CH

Khodinal. Menthone.

(i.) (ii.)

1 Bouveault and Gourmand, C. R., 1901, 138, 1699.

2 Tiemann, Per., 1898, 31, 2902.

102 RECENT ADVANCES IN ORGANIC CHEMISTRY

5. Citral.

By distilling together the calcium salts of formic and
geranic acids we obtain the aldehyde citral.1 Since this is a
general reaction, the constitution of citral svould probably be
that shown in the equation below —

CH,

(CH3)2C:CH.CH2.CH2.C:CH.COO-ca
H.COO— ca

I CH,

2.C:CH.COO-ca

= (CH3)20 : CH.CH2.CH2.0 : CH.CHO + CaC03

In support of this formula we may quote the decomposition
of citral into acetaldehyde and methyl-heptenone, which takes
place when the substance is warmed with a solution of sodium
carbonate.

Citral, therefore, represents rhodinal or citronellal, from
which two hydrogen atoms have been withdrawn ; and differs
from them further in that it contains no asymmetric carbon
atom. But though it loses this possibility of isomerism, it
retains another, for it has been found to occur in two
geometrically isomeric forms 2 —

H— C— CHO

(CH3)2C : CH . CH2 . CH2— C— CH3
Citral a.

CHO— C— H

II
(CH3)2C : CH . CH2 . CH2— C— CH3

Citral b.

These have been shown by Harries and Himmelmann to
be structurally identical ; and the relative configurations have
been deduced from the relations of the two compounds to
geraniol and nerol, with which we shall deal later.

Like the other olefinic terpenes, citral can be converted into
cyclic substances with great ease. When it is boiled for a long
time with glacial acetic acid it is changed into cymene 3 —

1 Tiemann, Ber., 1898, 31, 827, 2899.

2 Tiemann, Ber., 1899, 32, 115; 1900, 33, 877; Bonveault, Ball. soc. chim.,
1899, [iii.] 21, 419, 423; Barbier, ibid., 635; Kerschbaum, Ber., 1900, 33, 886;
Zeitschel, Ber., 1906, 39, 1783; Harries and Himmelmann, Ber., 1907, 40, 2823.

1 Tiemann and Semmler, Ber., 1895, 28, 2134.

THE OLEFINIC TERPENES
CH3 CH3

103

CH3 CH3

H2C CHOH

H2C CH2

\/
C— OH

CH,

CH3

Cymene.

A second condensation of citral takes place when the
aldehyde group is so treated that it takes no part in the
action. For instance, if we condense citral with a primary
amine, we obtain a cyclo-citral by a simple wandering of bonds
and ring-formation —

CH3 CH3

\/ (

c /

'/ H2C
CH.CH:N.R

II -^

3-OH
CH2.CH:Isr.E

PTT -s

il —r
H,C C . CH3 + 2H2o

\ / H2C
CH2

Citral derivative. C

(i.)

v^Jtl3 s

/ - 2H20

H2 OH
(II.)

104 RECENT ADVANCES IN ORGANIC CHEMISTRY

CH, CH, CH8 CH3

H2C CH.CHrN.R H2C CH.CHO

CH.CHrN.R H2C CH.

H2C C . CH3 H2C C . CH3

\ ^ \J*

CH CH

a-Cyclocitral.
(III.) (IV.)

The same result may be obtained by condensing citral with
cyan-acetic ester instead of an amine. In each case, the amine
or cyan-ester can be split off after the condensation to cyclo-
citral has taken place.

Cyclo-citral occurs in two isomeric forms,1 the formation of
either being dependent upon the manner in which water is
eliminated from the molecule of an intermediate hydratioii
product (!!.)• The formation of /3-cyclocitral takes place as
shown below.

CH3 CH3

C— OH

H2C CH2CH:N.R

/ \

H2C O.CHiKR

H2C C.CHO

CH3 3

> 1 1] >

1 II

H2C C.CH3

H2C C

\/\

\ /

CH2

CH2 OH

j8-Cyclocitral.

(II.)

The practical interest of citral lies in the fact that when it
is condensed with acetone by means of baryta, it yields a
substance, pseudo-ionone, which, by the action of sulphuric
acid, is changed into ionone,2 the basis of artificial violet
perfume —

1 Tiemann, Ber., 1900, 33, 3719.

2 Tiemann and Kruger, Ber., 1893, 26, 2691 ; Tiemann, ibid., 1898, 31, 808,
867, 1736, 2313 ; 1899, 32, 827 ; Tiemann and Schmidt, {bid., 1900, 33, 3703.

THE OLEFINIC TERPENES

105

ft
o

ft
Q

§-§ I

W
o

t__ J

N**>'

0 3

.........

LQ..

W
o

ft
o

^ft

W
o

io6 RECENT ADVANCES IN ORGANIC CHEMISTRY

This body differs from the natural substance irone (to which
the odour of violets is due) only in the position of a double
bond- CH

HC CH . CH : CH . CO . CH3
HC CH . CH3

CH2

Irone.

6. Geraniol, Nerol, and Linalool.

If we reduce citral with sodium amalgam in an alcoholic
solution weakly acidified with acetic acid, a mixture of two
isomeric alcohols, geraniol and nerol, is obtained. These two
bodies, on oxidation, regenerate citral, and on this ground, as
well as on account of other reactions common to both, it is
assumed that they are structurally identical but stereoisomeric
substances of the formula — CH3

(CH3)2C : CH . CH2 . CH2 . C : CH . CH2OH

Proof of the correctness of this formula is afforded by the fact
that when geraniol is heated with water to 150° C. it gives
ethyl alcohol and methylheptenone ; while on oxidation it gives
acetone, laevulinic acid, and oxalic acid.

By the action of acetic acid, to which one or two per cent,
of sulphuric acid has been added, both nerol and geraniol give
terpineol —

CH3 CH3 CH3

I I I
C C C

/ \ / \ / \.

H2C CH H2C CH H2C CH

H2C CH2OH H2C CH2OH H2C CH2

CH CH2 CH

II I I

C C— OH C— OH

CH3 CH3 CH3 CH3 CH3 CH3

Geraniol. Hypothetical glycol. Terpineol.

THE OLEFINIC TERPENES 107

Now, this reaction takes place nine times faster with nerol
than with geraniol ; and if the two bodies are geometrical isomers,
this difference allows us to draw a conclusion with regard to their
configurations.1 A comparison of the two formulas below will
suffice to show that in (I.) the groups which unite to form the
terpineol ring are further apart in space than they are in (II.).
The ring-formation will therefore occur more easily in the case
of (II.) than in that of (I.). Hence we must ascribe to geraniol
the first formula, and to nerol the second —

H— C— CH2OH

II
(0x13)20 '. OH . OH2 . OH2 — 0 — CH3

Geraniol.

(I-)
CH2OH— C— H

(CH3)2C : CH . CH2 . CH2— C— CH3
Nerol.
(II.)

We are now able to deal with the space formulae of the two
citrals. The oxidation of geraniol gives a mixture of citral a and
citral &, in which citral a predominates ; while with nerol the
proportions are reversed, more citral b being formed. From this
we may deduce that citral a has the same configuration as
geraniol, while citral b has its groups arranged as in nerol —

H— C— CHO

(CH3)2C : CH . CH2 . CH2— C— CH3
Citral a (Geranial).

CHO— C— H

(CH8)2C : CH . CH2 . CH2— C— CH3

Citral 6 (Neral).

Both geraniol and nerol are found in nature as inactive
substances, which agrees with the formulas which we have
ascribed to them above. The isomeric compound, linalool,
however, occurs in both dextro- and Isevo-rotatory forms, and
must therefore contain an asymmetric carbon atom. The
inactive form of linalool is convertible into both geraniol and
nerol by the action of acetic anhydride. This reaction can be
explained by assuming that linalool has the formula —

1 Zeitschel, Ber., 1906, 39, 1780.

io8 RECENT ADVANCES IN ORGANIC CHEMISTRY

(01X3)20 : OH . CH2 « CH2 . 0 . CH : CH2

I
OH3

A comparison of the formulae of geraniol, nerol, and this
one proposed for linalool will show that by the addition of
water to each of these substances we can produce in all three
cases the same glycol of the formula—

CH3

(CH3)2C : CH . CH2 . CH2 . 0 . CH2 . CH2OH

This formation of a common hydration product suffices to
explain the interconvertibility of the three isomers ; but there
is one point which seems to render the linalool formula rather
doubtful. When we take laevo-linalool and treat it with acetic
anhydride, terpineol is formed along with nerol and geraniol ;
and this terpineol is found to be dextro-rotatory. But when
we compare the formulae of terpineol and linalool, we find that
the asymmetric carbon atom of linalool does not correspond to
that in terpineol; in fact, the atom which in linalool was
asymmetric is now not asymmetric, while a new asymmetric
carbon atom has come into being. How optical activity can
persist through such a change as this appears difficult to under-
stand, unless we assume that it is a case of asymmetric synthesis
similar to those described later in this volume.

CH3 CH3 CH, CH3 CH3 CH2

C— OH C— OH C— OH

CH CH

/*\ /*\ _

CH2

H3C

H2C CH2 H2C CH

X v

CH3 OH CH3

Intermediate products. d-Terpineol.

This terminates our survey of the terpene class. In con-
clusion, we may append to this chapter a table showing some of
the possible conversions of mono-cyclic, di-cyclic, and olefinic ter-
penes into each other, and also into members of the benzene series.

THE OLEFINIC TERPENES

109

5 Hydrochloric
r" and aniline c

CHAPTEE VI

THE ALKALOIDS

A. — GENERAL

WHEN we attempt to define what we mean by the term
" alkaloid " our difficulties are not small. On the one hand,
our definition may be so drawn as to include almost every
naturally occurring nitrogen compound, which is obviously
useless as a mode of classification ; or it may be so narrow as to
exclude some of the most important of the substances which are
usually included in the alkaloid group. The most general
definition is perhaps the best ; and for our present purpose we
shall treat as alkaloids those naturally occurring substances
which contain cyclic chains, of which at least one member is a
nitrogen atom. This definition opens to us a much wider field
than we can possibly attempt to cover in the space at our
disposal, and in the following pages we shall aim at describing
the syntheses and constitutions of a few typical compounds
rather than at a survey of the whole subject.

Practically all the important alkaloids are found in the
tissues of vegetables ; and if we except xan thine derivatives, we
might have modified the definition given above by limiting the
\ term " alkaloid " to basic substances found in plants. )

The known members of the alkaloid class are very numerous,
and the number of workers in the field has been great ; this was
to be expected from the pharmacological importance of these
substances, which renders a knowledge of their structure of the
utmost value.

As the following pages will show, the chemistry of the
alkaloids resembles that of the aromatic compounds, in that
both classes seem to be built up upon the basis of one
substance. In the aromatic series benzene lies at the root of all
the compounds however complicated they be; while in the

THE ALKALOIDS . in

alkaloids pyridine appears to be equally essential And just as
among the aromatic types we find a benzene ring condensed
with other cyclic chains, so in the alkaloids we may discover
compounds in which the pyridine ring is overlaid with others.
Even the derivatives of the purine group may be considered to
be derived from pyridine by the substitution of a second nitrogen
atom in the ring.

According to Guareschi,1 the alkaloids are the degradation
products of protoplasmic action in plants. They do not seem
to be again assimilated by the plant once they are formed, but
remain in the saps in the same way as uric acid may remain in
the human tissues. Pictet 2 has dealt with the subject in some
detail, and we may here summarize his views.

In the first place, he believes that alkaloids are not produced
in plants by direct syntheses, but are rather to be regarded as
the decomposition products of much more complicated sub-
stances. But as soon as the alkaloid is formed in the plant, it
immediately reacts with some other plant product to form a
derivative. For example, some alkaloids, such as soline, are
glucosides as well as alkaloids ; so that it is probable that in their
case the first-formed alkaloid reacts with glucose within the
plant-tissues. A more common case, however, is that in which
the alkaloid condenses with an organic acid, as in the case of
cocaine or atropine. But by far the most common case of all is
that in which the alkaloid reacts with an alcoholic radical,
usually methyl alcohol, to form an ether. In this class of
derivatives the action of formaldehyde apparently lies at the
root of the syntheses. Alkaloids which contain a pyrrol ring
are probably derived from proteins ; and it is noteworthy that
while on the one hand Fischer has shown that albumen on
hydrolysis gives pyrrol derivatives, it has been proved by
Nencki, Koster, Zaleski, and Marchlewski that the same
nucleus is to be found in haemoglobin and chlorophyll.

With regard to the occurrence of the alkaloids in nature,
very little generalization is possible. The monocotyledons
seem to be the richest in members whose tissues produce these
substances ; while among the cryptogamia there appears to be

1 Guareschi, " Alkaloide," p. 414.

2 Pictet, Arch. soc. phys. nat. Geneve, 1905, IV., 19, 329; Arch. d. Pharm.,
1906, 244, 389,

ii2 RECENT ADVANCES IN ORGANIC CHEMISTRY

no alkaloid formation. Just as little regularity is found with
regard to the distribution of the alkaloids in the various
portions of the plants themselves. Though no general rule can
be formulated, it seems probable that alkaloids are most often
found in the fruits and sap, or, in trees, in the bark.

Since in most cases alkaloids occur as salts, they are
obtained from the actual plant tissues by the action of alkali,
which liberates the basic part of the molecule. If this be
volatile in steam, the alkaloid is obtained in this way ; but if it
be not thus volatile it is extracted from the tissues by treating
them with acids, which dissolve the alkaloids, forming solutions
of their salts, from which the free alkaloid is obtained by the
action of alkali. Final purification is carried out by crystalli-
zation of the alkaloid or of its salts. When extraction is
carried out on a small scale, chloroform is often used to remove
alkaloids from the tissues in which they occur.

The majority of alkaloids are solid substances, but one or
two are liquids which can be distilled without decomposition.
Nearly all of them have powerful actions upon the animal
organism ; but owing to our ignorance of the relation between
chemical constitution and physiological action, not much can
be said on the subject. In most cases alkaloids are found to
possess laevo-rotation, and it is very seldom that both optically
active forms are found in nature.

The alkaloids are usually classed according to the hetero-
cyclic rings from which they are built up. Thus • we have
the pyridine alkaloids, the quinoline alkaloids, and so forth.
For our present purposes this system of classification is very
suitable, and we shall therefore deal with the subject under
the following heads : —

The Pyridine Group.
The Pyrrolidine Group.
The Quinoline Group.
The Isoquinoline Group.
The Purine Group. f

There is another series of substances, the morpholine or
phenanthrene group, but the constitutions of its members are at
present undetermined, so we shall omit it from our survey.

THE ALKALOIDS 113

B. — METHODS EMPLOYED IN THE DETERMINATION OF
ALKALOID CONSTITUTIONS

After we have carried out an elementary analysis of an
alkaloid we are in a position to state its percentage composi-
tion, and by a molecular weight determination we can estimate
the number of atoms which its molecule contains. The next
step is the determination of the mode in which these atoms
are linked together in the alkaloid molecule, and we shall now
give a brief account of some common reactions which are
employed to solve this problem.

In the first place, since many alkaloids are known to be
esters, it is usual to employ some hydrolytic method in order to
see whether or not the alkaloid molecule can be decomposed
into some simpler grouping. To this end, the alkaloid may be
heated with water, acids, or alkalis until it is decomposed into
its component acid and base. This method, while breaking
up any salt or ester, does not, except in a few cases, result in
any further destruction of the structure of the body ; so that
from the constitutions of the two halves we are able to deduce
the constitution of the parent substance.

This method of decomposition, however, may not carry us
far enough, and it is usually seconded by a more violent action.
For instance, the alkaloid may be fused with alkali, distilled
with zinc dust, heated with bromine or phosphoric acid. When
reagents such as these are employed, the less durable part of
the molecule is usually shattered ; and in the reaction-product
we find some stable nucleus such as pyridine, quinoline, or
isoquinoline, from which the whole alkaloid is derived.

Again, many alkaloids exist in the form of methyl ethers.
These can be broken up by boiling with hydriodic acid (Zeisel's
method); and by passing the methyl iodide thus formed into
silver nitrate solution the number of methyl radicals split off
by the hydriodic acid may be estimated, and thus the number
of methoxyl groups in the alkaloid can be ascertained.

When the alkaloid contains an oxygen atom, it is of
importance to determine whether this occurs in a carbonyl,
carboxyl, hydroxyl, or ether group. The first is determined in
the usual way by the action of phenylhydrazine or hydroxyl-
amine; the hydroxyl group can usually be detected by

ii4 RECENT ADVANCES IN ORGANIC CHEMISTRY

acylating it or by the action of dehydrating agents, which
split off water and leave an unsaturated substance; while if
the alkaloid is an alkyl ether it can often be decomposed by
Zeisel's method. If the carboxyl group occurs in the alkaloid
under examination, there is not much difficulty in detecting
its presence.

All alkaloids contain nitrogen, but it is necessary to dis-
cover in what way this nitrogen is linked with the rest of
the molecule. Herzig and Meyer have devised a method of
determination for methyl-imino groups which is very useful in
this branch of research. The hydriodides of bases in which a
methyl group is attached to nitrogen, when heated to about
300° C., split off methyl iodide, which can be estimated with
silver nitrate just as in the case of the methoxyl group. A
somewhat similar decomposition results in the reaction which
is usually termed "exhaustive methylation." Here, by the
action of methyl iodide and silver oxide, assisted by dry dis-
tillation, a cyclic nitrogen compound may be made to lose its
nitrogen atom with but little alteration in the rest of the
molecule. The formulae will make the process clear without
further explanation.

CH,
H2C' NCH2 CH3I H2C" XCH, Distill H.CTH

H,CX ^CH, AgaO H,CX sCH, -H,O H,C CH,

H2C XCH Distill

CHXCH

H3C CH,
10-^N— CH,

CH; XCH3

CH2CH2

N(CH3)3+ H20

The final stages in the constitution determination of any
alkaloid are usually those in which the oxidation products of
the substance are studied. We need not describe the actions

THE ALKALOIDS 115

of the various agents employed, as they are all well known.
The most useful are potassium permanganate, hydrogen
peroxide, dilute nitric acid, and chromic acid.

We must now proceed to the examination of the evidence
which has been collected with regard to the syntheses and
constitutions of some alkaloids.

C. — THE PYEIDINE GROUP
1. Coniine.

The first alkaloid with which we shall deal is the substance
coniine, which deserves the foremost place on two grounds : it
is the simplest member of the alkaloid class, and it is the
first alkaloid which has been completely synthesized from the
elements. The complete synthesis will be given in the case of
this substance, as it is of historical interest, but in the case of
the other synthetic compounds we must confine ourselves to
the later steps in the process.

By heating together carbon and sulphur we can produce
carbon disulphide, which, by the action of chlorine, is converted
into carbon tetrachloride. By heat, this can be changed into
perchlorethylene, C12.C : C.C12, and when this is acted on by
ozonized air it yields trichloracetic acid. Eeduction with
potassium amalgam in aqueous solution changes trichloracetic
acid into acetic acid. From this, acetone is obtained by
distillation of calcium acetate, and by reducing the acetone so
formed we can produce isopropyl alcohol. The action of zinc
chloride upon the alcohol gives propylene by dehydration, and
by the addition of chlorine we can then form propylene chloride
Propylene chloride and iodine chloride together yield tri-
f chlorohydrin, from which glycerine is obtained by heating the
trichloride to 160° with a large excess of water. Glycerine,
by dehydration, gives allyl alcohol; and this, in turn, allyl
bromide; from which, by the addition of hydrobromic acid,
we obtain trimethylene bromide. Eeplacing the bromine atoms
by cyanogen groups we produce glutaric nitrile, and this,
on reduction, gives us pentamethylene diamine. On dry
distillation, the hydrochloride loses ammonium chloride and
is converted into piperidine,1 from which pyridine* can be

1 Ladenburg, Ber., 1885, 18, 3100.

* Pyridine was obtained in a bimpler way by Kamsay (Ber., 1877, 10, 736)

ii6 RECENT ADVANCES IN ORGANIC CHEMISTRY

obtained by oxidation. Pyridine combines with methyl iodide,
and when the pyridinium methyl iodide thus produced is heated
to about 300° C. it suffers intramolecular change and is con-
verted into the hydriodide of a-picoline. Picoline, when heated
to a high temperature with paraldehyde, gives a-propenyl-
pyridine, which, on reduction, gives isoconiine.1 On further
heating to 300°, or boiling with solid potash, this is converted
into racemic coniine.2 To separate the right- and left-handed
forms, active tartaric acid is used, since this substance can also
be obtained synthetically and its two antipodes can be separated
from each other by means of the sodium ammonium salt with-
out the interposition of any naturally occurring optically active
substance. The formulae below give the steps which we have
mentioned —

C-fS2->CS2->CCl4->

C:C1

CH3

CH.OH

CH3

CH2
CH
CH«

CH2C1
CHC1

CH,

CH2OH CH2Br CH2Br

II
CH,

CH,

I
CH,

CH.

)H COOH
-> | -> CO •
Jla CH3 |
CH3

CH2C1

CH2OH

CHC1 ->
CH2C1

CHOH -
1
CH2OH

XN" CH2.
-> CH2

CH2.NH2

CH2Br CH2CN CH2.CH2.NH2

CH2 NH

CHg CH2

by passing a mixture of acetylene and hydrocyanic acid through a heated tube.
Since acetylene is produced by a carbon arc in a hydrogen atmosphere, and
hydrocyanic acid is formed by sparking a mixture of acetylene and nitrogen,
this forms a simpler synthesis from the elements.

1 Ladenburg, Ber., 1889, 22, 1403.

2 Ibid., 1906, 39, 2486.

THE ALKALOIDS 117

CH2

H2C CH2

I |

-CH:CH.CH3 -> H2C CH.CH2. CH2.CH3

N < " NH

(d + Z)-Coniine -> d- or Z-Coniine.

2. Piperine.

When the alkaloid piperine is boiled with alcoholic potash
it is decomposed into piperidine and piperic acid.1 The consti-
tution of piperidine is established by the Ladenburg synthesis
from pentamethylene diamine, which we mentioned in connec-
tion with the synthesis of coniine, as well as by the formation
of piperidine from pyridine, by reduction. We have, therefore,
only to determine the constitution of piperic acid in order to
establish the constitution of piperine.

The decomposition of piperine may be expressed in the
following way : —

C17H1903N + H20 = C6HuN + C12H1004
Piperine. Piperidine. Piperic acid.

Fittig, by the action of permanganate, oxidized piperic
acid to an aldehyde, piperonal,2 which has the composition
C7H502 . CHO. On further oxidation, piperonal is converted
into the corresponding acid, piperonylic acid, C7H502 . GOOH.
Now, this substance can be synthesized by the action of
methylene iodide upon protocatechuic acid in presence of
caustic potash, and therefore it must be the methylene ether
of that acid.

rrr II I - 2HI

±i

C H0n

\I HO— II J— COOH \0— 11 J— COOH

Protocatechuic acid. Piperonylic acid.

By subtracting the atoms in piperonylic acid from those
which make up piperic acid, we find a surplus of four carbon

1 Babo and Keller, /. pr. Ch., 1857, 72, 53.

2 Fittig and Kemsen, Annal&n, 1871, 159, 142.

ii8 RECENT ADVANCES IN ORGANIC CHEMISTRY

and four hydrogen atoms. This — C4H4 — must be so attached
to the benzene ring of piperonylic acid that on oxidation it dis-
appears entirely and does not give rise to a second carboxyl
group in the molecule. The only way in which this condition
can be fulfilled is by inserting the group — GJIt — between
the carboxyl group and the benzene ring of piperonylic acid.
Piperic acid would thus be represented by —

CH2

\0— II J— C4H4— COOH

When piperic acid is allowed to react with bromine, it takes
up four atoms of the halogen, thus showing that it contains
two double bonds. These double bonds must be in the side-
chain between the nucleus and the carboxyl group, hence we
may ascribe the following formula to piperic acid : —

-CH : CH— CH : CH— COOH

The synthesis of piperic acid may be carried out in the
following way. Synthetic protocatechuic aldehyde l was con-
verted by methylene iodide and potash into piperonal,2 which,
when warmed with acetaldehyde and very dilute alkali (Claisen's
reaction), forms piperonyl-acrolein —

— CH:CH— CHO

Piperonyl-acrolein.

1 Tiemann and Koppe, JBer., 1881, 14, 2015,

2 Wegscheider, Monatsh., 1893, 14, 382.

THE ALKALOIDS 119

When this acrolein derivative is heated for several hours with
sodium acetate and acetic anhydride it condenses with a molecule
of acetic acid (Perkin's reaction), and forms piperic acid 1 —

— CH : CH— CH : CH— COOH

By converting piperic acid into its chloride and heating the
latter with piperidine in benzene solution, piperine is formed. 2

CH2-CH2
CH2 /

6-4 J— CHiCH— CH:CH— CO.Cl + HN CH2

CH2—

Piperic acid chloride. Piperidine.

UCH2-CH2
/ \

— CH : CH— CH : CH— CO- N CHS

Piperine.

In this way the alkaloid can be synthesized, its constitution
being proved by the synthesis and further certified by the
decomposition reactions which we have mentioned.

3. Trigonelline.

This alkaloid has the composition CvH^NOa. It was dis-
covered by Jahns 3 in 1885 ; and in the following year its
constitution was proved by Hantzsch,4 who obtained it
unintentionally in the course of an examination of some
derivatives of nicotinic acid.

1 Ladenburg and Scholtz, Ber., 1894, 27, 2958.

2 Rugheimer, Ber., 1882, 15, 1390 ; Fittig and Remsen, Annalen, 1871, 159, 142.

3 Jahns; Ber., 1885, 18, 2518.

4 Hantzsch, Ber., 1886; 19, 31.

120 RECENT ADVANCES IN ORGANIC CHEMISTRY

Hantzsch treated nicotinic acid (I.) with caustic potash and
methyl iodide, obtaining the methyl ammonium iodide of
nicotinic methyl ester (II.). When this is acted on by silver
oxide the iodine atom is exchanged for a hydroxyl group, and
the compound (III.) is produced, which at once loses water and
is converted into a betaine (IV.). This synthetic body was
isomeric with trigonelline, and on comparing the two substances
Jahns l found them to be identical. Trigonelline is therefore
the methyl-betaine of nicotinic acid.

— COOH If ^— COOCH

Nicotinic acid.

(III.) (IV.)

>/\s.

-COOH |f ^— CO

CH3 OH CH3

Hydroxy-acid. Betaine

(Trigonelline).

D. — THE PYRROLIDINE GROUP
1. Nicotine.

The alkaloid nicotine stands in a position midway between
the pyridine and the pyrrolidine groups ; for, as will be shown
presently, it contains both a pyridine and a pyrrolidine
nucleus. It therefore forms a convenient bridge by which we
can pass from the consideration of the one class to the other.

Nicotine is a basic substance having the composition
CioHuNfr Its constitution has been established by means of
the following reactions : —

1 Jahns, Ber., 1887; 20, 2840.

THE ALKALOIDS 121

1. Mtric acid, chromic acid, or potassium permanganate
oxidize nicotine l to nicotinic acid —

— COOH

2. By the action of bromine upon nicotine, two derivatives 2
are formed —

(a) Dibromocotinine,

(b) Dibromoticonine,

3. When dibromocotinine is decomposed by bases it gives

methylamine, oxalic acid, and a compound C7H7NO.
By the same treatment dibromoticonine yields methyl-
amine, malonic and nicotinic acids.

4. Nicotine is a di-tertiary base,3 giving two isomeric

methyl iodide addition products.

From the first reaction, it is obvious that nicotine must be
pyridine, with a side-chain in the j3-position.

— C6HWN

From the third reaction it is clear that of the two nitrogen
atoms in nicotine, one carries a methyl group. This one
cannot be the pyridine nitrogen. Further, the second nitrogen
atom (which does carry the methyl radical) cannot belong to
a pyridine ring. We may thus go a step further, and represent
nicotine by the formula —

— C5H7:N.CH3

1 Huber, Annalen, 1867, 141, 271 ; Weidel, Annalen, 1873, 165, 328; Laiblin,
Ber., 1877, 10, 2136.

2 Pinner, Per., 1893, 26, 292.

3 Pictet and Genequand, Ber., 1897, SO, 2117.

122 RECENT ADVANCES IN ORGANIC CHEMISTRY

Again, the third reaction shows us that dibromocotinine
and dibromoticonine give rise to three carbon chains —

i. r

C— CH2— C — C.C— ,

Malonic acid chain. Oxalic acid chain. C7H7NO chain.

These must be somehow combined in the nicotine molecule, so
we may write the nicotine skeleton thus —

C— C

-C C

To this we must attach the group : N . CH3 in some way. From
the fourth reaction we deduce that this nitrogen atom is
a tertiary one, so that the two isomeric methyl iodide addition
products may be explained by the addition of methyl iodide
to a different nitrogen atom in each case. But if the group
: N . CH3 is to contain a tertiary nitrogen atom, and also to be
attached to the nicotine skeleton given above, the only way is
to make the nitrogen atom a member of a ring. The 'con-
stitution of nicotine would then be—

OJtl

The synthetic preparation of nicotine proved to be a much
harder task than was anticipated. The first steps were taken
by Pictet and Crdpieux,1 who, by heating /3-amido-pyridine (I.)
with mucic acid, were able to produce (II.) N-/3-pyridyl-pyrrol.
Like many other N-alkyl derivatives of pyrrol, this substance
when passed through a heated tube undergoes a molecular
1 Pictet and Crepieux, Per., 1895, 28, 1904.

THE ALKALOIDS 123

rearrangement, in the course of which the pyridine group is
transferred to the carbon atom next the nitrogen in the pyrrol
ring. The compound thus formed is a/3-pyridyl-pyrrol (III.).

(III.)
HC— CH

/^ / /^

NH,

0-amido-pyridine. N-£-pyridyl-pyrrol. 0-pyridyl-a-pyrrol.

From this substance Pictet 1 continued the synthesis in the
following way. The a)3-pyridyl-pyrrol forms a potassium salt,
the imino-hydrogen of the pyrrol group being replaced in the
usual way by the metallic atom; and from this salt, by the
action of methyl iodide, we obtain the methyl derivative of
the iodomethylate (IV.). On distillation with lime, this forms
the base nicotyrine (V.).

(IV.) (V.)

HC— CH HC— CH

C ^H

Nicotyrine.

Now, this body cannot be reduced direct to nicotine, for any
agent which attacks the pyrrol nucleus will, at the same time,
reduce the pyridine ring. The transformation can be carried
out in the following way, however. The nicotyrine (V.) is
treated with iodine in alkaline solution, by which means a
mono-iodine derivative is produced ; it in turn is acted on by
tin and hydrochloric acid, whereby it is partially reduced,
forming dihydro-nicotyrine (VI.). This substance reacts with
bromine to form a perbromide, C5H4N . C5H8N . Br4, which,
by reduction with tin and hydrochloric acid, yields inactive
nicotine (VII.). This racemic base can, like coniine, be resolved

> Pictet, C. R., 1903, 137, 860.

124 RECENT ADVANCES IN ORGANIC CHEMISTRY

into its antipodes by means of tartaric acid; so that in this
way the synthesis of Isevo-nicotine, corresponding to the natural
alkaloid, can be accomplished.

(VI)
HC— CHa

CH5

CH3

Dihy dronicoty rin e.

(VII.)
CH2 —

— CH CH

N
OH

Nicotine.

2. Tropidine.

Hitherto in this chapter we have confined our attention to
substances which contain a single ring of carbon and nitrogen
atoms; but with the tropine series we enter a new class in
which we shall have to deal with bridged rings analogous to
those of the dicyclic terpenes. The first member of the group
we are about to examine is tropidine.

Willstatter has succeeded in synthesizing this body in two
ways,1 one of which we may describe. The complete synthesis
is made up of two distinct stages, in the first of which cyclo-
heptene is converted into cycloheptatriene ; the second stage
deals with the formation of the nitrogen bridge across the seven-
membered carbon ring.

Suberone, the starting material, is obtainable by the distil-
lation of the calcium salt of suberic acid. It can be converted,
by reduction, into suberyl alcohol, and hence into suberyl
iodide, which, by the action of potash, may be made to lose
hydriodic acid and yield cycloheptene.

(I.)
-CH2— CO

I
CH,

(II.)
CH2— CH2— CH.OH

CH2

CH2 — CH2 — CH2

Suberyl alcohol.

1 Willstatter, Annalen, 1901,317, 268; 1903, 326, 1.

CH2 — CH2 — CH2

Suberone.

THE ALKALOIDS

125

(III.)
CH2— CH2— CH . I

CH2 CH2 CH2

Subervl iodide.

(IV.)
CH2— CH2— CH

II
CH

CH2 — CH2 — CH2

Cycloheptene.

Bromine is now allowed to act upon this, forming the
dibromide (V.); from which, by the action of two molecules
of dimethylamine, hydrobromic acid is removed, a dimethyl-
amine group being attached to the ring at the same time.
The substance (VI.) is thus formed.

(V.)
CH2— CH2

CH.Br
CH . Br

(VI)
CH2— CH2— CH . N(CH3)2 . HBr

2 —

CH2— CH

This is subjected to the action of methyl iodide, and the
addition compound thus formed is converted into a hydroxide of
the ammonium base. On distillation, this body splits off water
and trimethylamine, breaking down into cycloheptadiene (VII.).
By a repetition of the same process, a third double bond is
inserted in the ring ; or the same result may be attained by
adding two atoms of bromine to (VII.) and splitting off two
molecules of hydrobromic acid by means of quinoline. In any
case the resulting compound has the constitution of (VIII.).

(VII.)
CH2— CH=CH

II
CH2— CH2 — CH

Cycloheptadiene.

(VIII.)
CH2— CH=CH

CH=CH-CH

Cycloheptatriene. ,

Having thus obtained cycloheptatriene, we must turn to
the second stage in the synthesis and examine the means by
which the ring is bridged. By the action of one molecule of
hydrobromic acid upon the compound (VIII.) we obtain the
monohydrobromide (IX.), which reacts with dimethylamine at

126 RECENT ADVANCES IN ORGANIC CHEMISTRY

ordinary temperatures to give dimethylamino-cycloheptadiene,
which is identical with a-methyl-tropidine (X.).

(IX.)

CH2— CHBr— CH
CH

CH2-CH==CH

Monohydrobromide of
Cycloheptatriene.

(X.)
N(CH8),

CHa— CH— CH

II
CH

I
2 — =C-H.

a-Methyl-tropidine.

When acted on by sodium in alcoholic solution, this methyl-
tropidine takes up two hydrogen atoms, and is converted into
a-methyl-tropane (XI.). The action of bromine in acid solu-
tion gives a dibromide (XII.), which, on heating, undergoes
intramolecular change into bromotropane-methylammonium
bromide (XIII.).

N(CH8)2

CH2— CH— CH2

I
CH2

CH2— CH=CH
(XL)

o-Methyl-tropane.

CH2— CH-

CH2— CH

-CH2
CH2

CHa— CHBr— CHBr
(XII.)

Dibromide.

CH2

CH3

]

sT— CH3 C

\ i

CH2— CH CHBr

(XIII.)

By the action of caustic potash upon this last substance,
hydrobromic acid is split off and the methyl bromide addition
product of tropidine (XIV.) remains, which is converted into

THE ALKALOIDS

127

the chloride and then dry distilled. Tropidine (XV.) is thus
produced.

CH2— CH OIL

CH2— CH

-CH,

N— CH3 CH

CH.-CH-

-CH

(XIV.)

N— CH3 CH

I If

CH2— CH CH

Tropidine.
(XV.)

This synthesis leaves very little doubt as to the constitu-
tion of tropidine. At first sight it might be supposed that if
the intramolecular change which converts (XII.) into (XIII.)
were to involve the other bromine atom instead of the one
chosen above, a different product would be obtained —

/CH3

Br— 1ST— CH3

CH5

CHB

Examination will show, however, that this is identical
with (XIII.).

3. Tr opine, ^-Tropine* and Tropinone.

We must now consider the question of the synthesis of
tropine from tropidine. This cannot be directly accomplished,
but is attained through an intermediate product, i//- tropine,
which is stereo-isomeric with tropine. The method is as
follows.1 Tropidine is heated with hydrobromic acid in acetic
acid solution, by which means a-bromotropidine hydrobromide
(I.) is obtained. When the solution of this substance is
treated with ammonia or caustic alkali, bromotropane (II.) is

* The Greek ^ is used instead of the word " pseudo." Thus ^-tropine
represents pseudo-tropine.

1 Willstatter, Ber.t 1901, 34, 3163; Anndlen, 1903, 326, 23; cf. Einhorn,
Ber., 1891, 23, 2889.

128 RECENT ADVANCES IN ORGANIC CHEMISTRY

precipitated. On heating this with dilute sulphuric acid above
200° C., the bromine atom is replaced by a hydroxyl group and
^-tropine (III.) results.

(I.)

'1TT pTT

(

yH2

._AL12 v-'JLL— ~

CH.Br

(II.)

CH2— CH—

-CH,

OJtl2 OH Oxl2

Bromotropidinehydrobromide.

CH2—

N.CH3 CH.Br
CH2— CH CH2

Bromotropane.

(in.)

-CH,

N.CHo CH.OH

CH2— CH-

-CH,

^/-Tropine.

The isomerism of tropine and ^/-tropine may be explained
very simply. If the space formula of a compound having the
constitution of tropine be built up, it will be found that there
are two possibilities : the hydroxyl and the methyl groups may
lie on the same side of the ring as in (A), or on opposite sides
as in (B)—

(A) (B)
CH2— OH CH2 CH2— CH CH2

CH3

/ c

H
CH2— CH CH,

CHc

HO
CH2 — CH CH2

Now, of the two, tropine is the labile isomer, so that while we
can convert it at will into ^/-tropine, the reverse change is not
possible direct. Willstatter and Iglauer,1 however, have been
able to obtain tropine from ^-tropine by an indirect method.
They oxidize ^-tropine to tropinone —

1 Willstatter and Iglauer, Her., 1900, 33, 1170.

THE ALKALOIDS 129

2 — OH -- CH-?

I I

N . CH3 CO

I I

CH2— CH -- CH2

Tropinone.

and from this ketone they obtain tropine itself by the action of
zinc dust and concentrated hydriodic acid —

CH2— CH

N.CH3
CH2

CH2

CH . OH

CH -- CH 2

Tropine.

It will be noticed that in the foregoing paragraphs we
neglected to take into account a possible alternative formula
for bromo-tropane, which, if correct, would invalidate our
conclusions with regard to the constitutions of tropine and
tropinone. The formula of tropidine is given below, and it will
be seen that hydrobromic acid might be added on to it in either
of two ways —

/-ITT r^TJ

V^-ll2 V^Jtl

-CH2

V ^7

N.CH3

CH.Br(L)

^R PH PH ^ ^^'

])H2 — CH

1
-CH2

jL\% vyiJ.-111 '•- V_/iA2 ^^ —

1 1 ^ (

N.CH3 CH

1 II \#a, (

JH2— CH CH <^

^H PH

CH2

Tropidine. "^

CH2 (II.)

^TT_ P.TT

PTT T^v

Now, the tropinone derived from formula (II.) would have
the following constitution —

_CH CH2

N.CH3 CH2
OH,— CH CO

130 RECENT ADVANCES IN ORGANIC CHEMISTRY

which differs from the tropinone we have already dealt with
(derived from (I.)) in having only one methylene group next
the carbonyl, while the other tropinone has two methylene
groups, one on each side of its carbonyl radical. But from an
examination of the properties1 of tropinone obtained from
bromo-tropane, as already described, it is found that it must
have two methylene groups adjacent to its carbonyl radical.
For example, it forms a ^-isonitroso-compound with nitrous
acid ; benzaldehyde condenses with it to form a d^-benzal
compound ; while with oxalic ester it gives tropinone-^-oxalic
ester. The presence of the group — CH2 — CO — CH2 — in
tropinone is thus established, which at once disproves the
possibility that tropinone is derived from a bromo-tropane of
formula (II.).

4. Tropic Acid.

By the synthesis of tropine we have approached that of
another alkaloid, atropine. This substance, when boiled with
baryta water, breaks down into tropine and tropic acid. We
have thus established the constitution of half the atropine
molecule; and in the present section we shall deal with the
constitution of the other half.

Tropic acid has been synthesized by Ladenburg and
Eiigheimer.2 Acetophenone is treated with pentachloride of
phosphorus, whereby the oxygen atom is replaced by two
chlorine ones, and acetophenone chloride is formed. This is
allowed to react with potassium cyanide in alcoholic solution
to form the nitrile of atrolactinic ethyl ether —

CH3
C6H5— C— OEt

The nitrile is then hydrolyzed, forming the acid. When this
body is boiled with concentrated hydrochloric acid it loses
alcohol, and is converted into atropic acid —

1 Willstatter, Per., 1897, 30, 2679.

2 Ladenburg and Rugheimer, Ber., 1880, 13, 376, 2041.

THE ALKALOIDS 131

CH2

C6H5— C— COOH

Hydrochloric acid then attaches itself to the double bond,
yielding ]3-hydrochloratropic acid —

CH2C1
C6H5— OH . COOH

This substance, when boiled with potassium carbonate,
exchanges a chlorine atom for a hydroxyl group, and is
converted into tropic acid —

CH2OH

-CH— i

C6H5— CH— COOH

Tropic acid.

5. Atropine.

The constitutions of the two halves of the atropine molecule
have now been established, and the atropine synthesis can be
carried out by treating a mixture of tropine and tropic acid
with hydrochloric acid gas in the usual way.1 Atropine,
therefore, is the tropine ester of tropic acid, and it must have
the constitution shown by the following formula : —

CH2— CH CH2 CH2OH

N.CH3 CH.O. CO.CH.C6H5

CH2— CH— — CH2

Atropine.

The synthesis of atropine from the elements may be
accomplished in the following steps. Glycerine is obtained by
the reactions already described in the section on coniine, and
from it glutaric acid is produced. This body, by the electro-
lysis of the sodium salt of its mono-ester,2 gives suberic acid,
which is then converted into tropine by the method we have
described under that head. With regard to tropic acid, we

1 Ladenburg, Ber., 1879, 12, 941 ; 1880, 13, 104.

2 Crum Brown and J. Walker, Annalen, 1891, 261, 119.

132 RECENT ADVANCES IN ORGANIC CHEMISTRY

may start from acetylene, which can be produced by a carbon
arc in a hydrogen atmosphere. On the one hand, we can convert
the acetylene thus obtained into ethylene, ethyl- sulphuric acid,
alcohol, acetic acid, and, finally, acetyl chloride ; while, on the
other hand, we can polymerize it direct to benzene by passing
it through a red-hot tube. From the acetyl chloride and
benzene we can produce acetophenone by the Friedel- Crafts'
reaction, after which we proceed as already described under
tropic acid.

6. Ecgonine.

Tropinone forms salts with alkalis, and these, by treatment
with carbonic acid in the usual way, can be converted into the
alkali salts of carboxylic acids.1 In the case of the sodium
salt, it is suspended in ether, and carbon dioxide passed through
the liquid at ordinary temperatures; the resulting product is
the sodium salt of tropinone carboxylic acid, and when this is
reduced with sodium amalgam in a weakly acid solution it
yields a mixture of two isomeric bodies having the same
composition as ecgonine, C8Hi4NO . CO OH.

The two isomers, however, differ in character. The one
has all the properties of ecgonine, except the power of rotating
the plane of polarization ; it is a true carboxylic acid, forming
salts and esters, it also possesses a free hydroxyl group, and can
be converted into esters by acids. The second isomer, on the
other hand, behaves quite differently. It possesses no free
hydroxyl group, nor can it be esterified by the ordinary
methods. An explanation of the formation of two such
substances is to be found by considering the character of the
sodium derivative of tropinone.

It is well known that the sodium salts of ketonic bodies
usually exist in the enolic form, so that we should incline to
write the formula of the tropinone sodium salt thus —

CH2— CH CH

I II

N— CH3 C— 0— Na

CH2— CH— -CH2

1 Willstatter and Bode, Ber., 1900, 33, 411.

THE ALKALOIDS 133

The action of carbon dioxide upon this would produce a
sodium salt whose constitution could be written —

CH2— CH CH

I II

N.CH3 C— 0— COONa

CH2— CH- -CH2

This body forms by far the greater proportion of the reaction
mixture, but since the sodium salt of tropinone exists in the
keto- as well as in the enol-form, part of the end-product will
have the constitution shown below —

CH2— CH CH . Na CH2— CH CH . COONa

2— CH -- CH

CH3 CO

CH2 — CH CH2 CH2 — CH CH2

This last substance, on reduction, would give us the
alcohol —

CH2— CH CH . COOH

N.CH3 CH.OH
— OH Cli2

which proves to be racernic ecgonine.

7, Cocaine.

From ecgonine, cocaine can be prepared by benzoylating the
alcohol radical, and then esterifying the carboxyl group with
methyl alcohol.

CH2-CH- -CH . COOCH3

1ST . CH3 CH . 0 . CO . C6H5

CH2— CH- -CH2

Cocaine.

134 RECENT ADVANCES IN ORGANIC CHEMISTRY

E. — THE QUINOLINE GROUP
1. The Constitution of Cinchonine.

The alkaloid einchonine has the composition Ci9H220N2.
The oxygen atom forms part of a hydroxyl group, as is shown
by acetylation ; and the two nitrogen atoms are tertiary ones.

I. When einchonine is oxidized by means of chromic acid
and sulphuric acid l it breaks down into two substances,
cinchonic acid and meroquinene, in accordance with the
following equation : —

-f 30 = Ci0H702N + C9H1502N
Cinchonine. Cinchonic Meroquinene.

acid.

Cinchonic acid has been shown to be a quinoline carboxylic
acid of the formula —

COOH

so that einchonine itself must be a y-quinoline derivative.

For the sake of convenience, we will refer to the two halves
of the einchonine molecule as the " quinoline half " and the
" second half." It is obvious that the hydroxyl group which
is known to exist in the einchonine molecule must be situated
in the " second half" ; for if it were in the "quinoline half" it
would appear in cinchonic acid. We may therefore formulate
Cinchonine in the following way : —

C10H15(OH)N

Konigs, Ber., 1894, 27, 1501.

THE ALKALOIDS 135

II. Now, when cinchonine is oxidized with potassium
permanganate1 instead of chromic acid, the decomposition
products are quite different from those obtained before. The
reaction takes the course shown below —

Ci9H22ON2 + 40 = Ci8H2003N2 + H . COOH
Cinchonine. Cinchotenine.

This new oxidation product, cinchotenine, contains the quino-
line nucleus (as is shown by its behaviour on further oxidation).
It is therefore produced by a decomposition in. the " second
half" of the molecule. It contains a hydroxyl and a carboxyl
group. Cinchonine can take up one molecule of a halogen
acid, but cinchotenine has lost this property. Hence the group
CH2 of cinchonine has been split off, leaving the carboxyl
group in cinchotenine. We may thus carry our deductions a
step further, and write the formula of cinchonine in the follow-
ing way : —

/CH : CH2

III. We must now turn to a different reagent. When
cinchonine is treated with phosphorus pentachloride and then
with alcoholic potash it loses a molecule of water and is
converted into cinchene 2 —

C19H22ON2 - H20 = C19H20N2
Cinchonine. Cinchene.

When heated with twenty-five per cent, phosphoric acid,3
cinchene takes up two molecules of water and is decomposed
into lepidine and meroquinene —

Ci9H20N2 + 2H20 = C10H9N + C9H1502N
Cinchene. Lepidine. Meroquinene.

1 Konigs, Annalen, 1879, 197, 374.

2 Comstock and Konigs, Per., 1884, 17, 1985.

3 Konigs, Ber.. 1890, 23, 2677; 1894, 27, 900.

136 RECENT ADVANCES IN ORGANIC CHEMISTRY
Lepidine is known to have the formula —
OH,

IV. Meroquinene is the next substance whose decomposi-
tions must be examined. When it is oxidized with an ice-cold
mixture of sulphuric acid and potassium permanganate it gives
cincholoiponic acid 1 —

C9H1502lSr + 40 = C8H130*N + H . COOH

Meroquinene. Cincholoiponic

acid.

This, by the action of aqueous permanganate, is converted
into loiponic acid 2 —

[ + 02 = C7Hn04N + H . COOH

Cincholoiponic acid. Loiponic acid.

Loiponic acid is an unstable form of hexahydrocinchomeronic
acid, for on heating with caustic potash it is converted into that
substance by isomeric change. By assuming the structure of
loiponic acid to be the same as that of hexahydrocinchomeronic
acid (the configurations of the two being different), we can
work back step by step to meroquinene, whose formula must
therefore be that shown in the series below —

CH2.COOH CH2.COOH

I I

CH CH

H2C CH.CH:CH2 H2C CH . COOH

II II

H2C CH2 H2C CH2

v v

Meroquinene. Cincholoiponic acid.

1 Konigg, Ber., 1895, 28, 1986, 3150.

2 Skraup, Monatsh., 1896, 17, 377; Konigs, Ber., 1897, 80, 1329.

THE ALKALOIDS 137

COOH

H2C CH . COOH

Loiponic acid.

The position of the — CH2 . COOH group of meroquinene
is uncertain.

The formula above is due to Konigs, but the alternative put
forward by Miller and Rohde 1 —

CH3— C— COOH

H2C CH.CH:CH2 <,

I I
H2C CH2

\ /
NH

has probably as much to recommend it.

Of the ten carbon atoms of the "second half" we have thus
established the mode of linkage of eight: five in a piperidine
ring, two in a vinyl group, and one in a methyl or methylene
group. The ninth carbon atom of the "second half" must be
utilized in joining the two halves together. Thus we have only
to determine the position of the tenth carbon atom of the
"second half."

V. It will be remembered that the two nitrogen atoms
of cinchonine are tertiary; but it has been shown that the
nitrogen atom of meroquinene is a secondary one. This has
been established by the usual reactions of the imido-group, and
agrees with the constitution which we have ascribed to mero-
quinene in the previous paragraph. This peculiar behaviour
of the nitrogen atom can best be explained by the assumption
that in the "second half" of cinchonine we have a nucleus of
either of the types (I.) or (II.) —

1 Miller and Bolide, Ber., 1895, 28, 1060.

138 RECENT ADVANCES IN ORGANIC CHEMISTRY

Konigs' view.
CH

v. Miller's view.
CHfl

When such a nucleus as (I.) is heated with dilute acids it
will undergo intramolecular change into an imido-ketone in the
way expressed by the formula (la.) below. If the type (II.)
be chosen instead of (I.) the analogous substance (Ha.) would
be produced in the same way.

Such a change actually occurs when cinchonine is heated
with dilute acetic acid; an imido-ketone results, which, on
account of its poisonous properties, is named " cinchotoxine." l
Thus it is apparent that across the piperidine ring there is a
bridge of one carbon atom, and this accounts for the missing
tenth carbon atom in the " second half " of cinchonine.

From the foregoing evidence, cinchonine would be repre-
sented by either of the two formulae below —

1 Miller and Bolide, Ber., 1894, 27, 1187, 1279; 1895, 28, 1056.

THE ALKALOIDS

139

H.CH:CH2

2. 5^6 Constitution of Quinine.

Knowing the constitution of cinchonine, we can easily prove
that of quinine.

I. Quinine differs from cinchonine by one carbon, one
oxygen, and two hydrogen atoms —

CaoHaANa - Ci9H22ON2 = CH20
Quinine. Cinchonine.

This points to quinine being a methoxy-derivative of cincho-
nine, if we bear in mind the similarity in character between the
two substances.

II. When oxidized with sulphuric and chromic acids,1
quinine gives the acid (A) ; whereas it will be remembered that
cinchonine gave cinchonic acid (B). Meroquinene is one of
the oxidation products in both cases.

COOH COOH

— OCH3

(A) (B)

III. This proves the presence and position of the methoxyl

1 Skraup, Monatsh., 1881, 2, 591 ; 1883, 4, 695 ; 1891, 12, 1106 ; 1895, 16, 2684.

140 RECENT ADVANCES IN ORGANIC CHEMISTRY

group in quinine ; and since in its reactions quinine forms an
exceedingly close analogue to cinchonine, we are justified in con-
cluding that it is a methoxy-cinchonine of the following consti-
tution (accepting Konigs' view of the structure of cinchonine) : —

CH2 CH . CH : CH2
H2C HO— CK CH2

^T^

CH,

OCH<

Quinine.

3. CincJionidine and Gonchinine.

Cinchonine has three asymmetric carbon atoms in its mole-
cule, and therefore it may occur in several stereoisomeric
forms. Cinchonidine is supposed to be one of these ; while
conchinine is a stereoisomer of quinine.

F. — THE ISOQUINOLINE GROUP.
1. The Constitution of Papaverine.

The constitution of papaverine is a much simpler question
than that with which we have just dealt in the case of
cinchonine. There are six steps in the argument.1

I. In the first place, the formula of papaverine is
C20H2i04N"; it contains four methoxyl groups, which can be
hydrolyzed, yielding the substance papaveroline, Ci6H9N(OH)4.
This accounts for all the oxygen atoms.

II. On fusion with alkali, papaverine breaks down into two
nuclei, one of which contains nitrogen, while the other nucleus
is nitrogen-free. The first was proved to be a dimethoxy-
quinoline of the constitution —

1 Goldschmiedt, Monatsh., 1883, 4, 704 ; 1885, 6, 372, 667, 954 ; 1886, 7, 485 ;
1667, 8, 510; 1888, 9, 42, 327, 3i9, (579, 762, 778 ; 1889, 10, 673, 692.

THE ALKALOIDS

141

while the second decomposition product was dimethyl-
hoinocatechol —

CH«—

— OCHc
-OCH.

III. The fact that these two groups are directly united to one
another follows from the composition of the two decomposition
products —

H2 + CaoHaAN = CnHnOaN + C9H1202

Papa verm e. Dimethoxy- Dimethoxy-
quinoline. homocatechol.

IV. We must now examine the question of the manner in
which the two nuclei are united. Since papaverine contains
four methoxy-groups, and each of the decomposition products
contains two, it is obvious that during the decomposition no
methoxy-group is destroyed. Now, if the link between the
two nuclei had been an oxygen atom, i.e. if papaverine had
contained the grouping K — 0 — CH2 — 0 — E, then in the
breakdown of the molecule one — 0 . CH2 . 0 — group would have
been destroyed. We may therefore exclude the idea of joining
the two nuclei through an oxygen atom, and must assume that
they are directly united, carbon to carbon.

V. Our next problem is to find which carbon atom of the
isoquinoline ring is joined to the other nucleus. When we
oxidize papaverine with potassium permanganate, we obtain
a-carbocinchomeronic acid —

142 RECENT ADVANCES IN ORGANIC CHEMISTRY

Hence we deduce that the side-chain (second nucleus) was
attached at the point now occupied by the carboxyl group,
which is marked with an asterisk. Papaverine is therefore —

VI.

0,H6(OCH8)2

We have now to settle the constitution of the
group — C7H5(OCH3)2. This must be the dimethoxy-homo-
catechol radical, which has the same composition. We have
only to decide whether the two nuclei are joined ring to ring
or by the intermediation of the side-chain of the dimethoxy-
homocatechol. Without going into details, it may be said that
all the evidence points to the union being made through the
side-chain. The constitution of papaverine is therefore —

2. The Synthesis of Papaverine.

The synthesis of papaverine has recently been carried out by
Pictet and Gams.1 The reactions may be grouped in five

I. The first step in the process is the synthesis of amino-
aceto-veratrone. For this purpose veratrol (I.) is treated with

1 Pictet and Gams, G.E , 1909, 149, 210.

THE ALKALOIDS 143

acetyl chloride in presence of aluminium chloride, whereby
aceto-veratrone (II.) is formed. When this is treated with
sodium ethylate and amyl nitrate, it yields the isonitroso-
derivative (III.), which can then be reduced by tin chloride
and hydrochloric acid to the hydrochloride of amino-aceto-
veratrone (IV.).

CH30— CH30— X\- CO— CH3

CH30-

(III.) (IV.)

II. We must now turn to the synthesis of homoveratroyl
chloride. Vanillin (V.) is methylated and then treated with
hydrocyanic acid, giving dimethoxy-mandelic nitrile (VI.).
When this is boiled with hydriodic acid three processes take
place simultaneously; reduction, hydrolysis and the splitting
off of methyl radicals. We thus obtain homoprotocatechuic
acid (VII.) and by methylation of the hydroxyl groups
followed by the action of phosphorus pentachloride the chloride
of homoveratric acid is formed (VIII.).

CH30— /\,CHO CH30— / \CH(OH) .ON

HO— if ^— CHo— COOH

(VII.)

III. If we now allow the amino-aceto-veratrone hydro-
chloride obtained in Stage I. to interact with the homoveratric

144 RECENT ADVANCES IN ORGANIC CHEMISTRY

chloride of Stage II. in presence of alkali, we obtain homo-
veratroyl-amino-aceto-veratrone (IX.),

CH30-
CH30—

.CO.CH2.XH.CO.CH2.

(IX.)

IV. An inspection of the formula (IX.) will show that
though the substance contains two carbonyl groups, one of
these is a true carbonyl while the other is a radical which
originally formed part of a carboxyl group. When the
substance is reduced with sodium amalgam in neutral alcoholic
solution, the true carbonyl is reduced, while the acidic carbonyl
remains unaffected. The product is homoveratroyl-hydroxy-
homoveratrylamine (X.).

CH30

CH30-

. CH(OH) . CH2 . NH . CO . CH2 .

(X)

— OCH3

— OCH<

V. When this substance (X.) is treated with phosphorus
pentoxide in boiling xylene solution, it loses two molecules
of water and is converted into papaverine (XL).
OH

CHa -2H20 CH30-
CH30—

— OCH3

THE ALKALOIDS

3. The Synthesis of Laudanosine.

In the preceding section we have seen how the synthesis
of papaverine may be accomplished, and we are now in a
position to consider the question of a closely related alkaloid,
laudanosine. This body is very simply produced from papa-
verine. Pictet and Athanasescu 1 showed that if we form the
chloro-methyl derivative of papaverine and then reduce this
with tin and hydrochloric acid we obtain methyl-tetrahydro-
papaverine. This synthetic substance is of course racemic ;
and from it the dextro-antipode was obtained in the usual way
by making the quinic acid salt of the alkaloid and fractionally
crystallizing it. The substance thus obtained was found to be
identical with natural laudanosine.

— OCH3

OCH3

Papaverine.

>CH3

Laudanosine.

Pictet and Finkelstein 2 have recently carried out the
complete synthesis of laudanosine, but as the method is very
similar to that which we have already described in the case of
papaverine we need not enter into it here.

1 Pictet and Athanasescu, Ber., 1900, 33, 2346.

2 Pictet and Finkelstein, Ber., 1909, 42, 1979; C. .R., 1909, 148, 925.

146 RECENT ADVANCES IN ORGANIC CHEMISTRY

4. Opianic Acid.

Though opianic acid itself is not an alkaloid, we must take
up its constitution at this point owing to its relation with
narcotine, with which we shall deal later.

I. When narcotine is hydrolyzed with barium hydrate
or sulphuric acid,1 it decomposes into opianic acid and
hydrocotarnine —

G22H23U7^N -}- H2O — CioHioOs -f- Oi2Hi5(J3^I
Narcotine. Opianic acid. Hydrocotarnine.

II. Opianic acid is a monobasic acid, and therefore we may
write its formula, C9H903 . COOH.

III. When heated with hydriodic acid, two methyl groups arc
split off from opianic acid.2 It therefore contains two methoxy-
groups, and may be written thus, (CH3O)2 . C7H80 . COOH.

IV. When heated with potash3 it gives (by reduction)
meconine, and (by oxidation) hemipinic acid —

OCH3 OCH3

CH30— f\— CO v CH30— /\—

M>H,

Meconine. Hemipinic acid.

This last reaction is parallel to the formation of benzyl alcohol
and benzoic acid by the action of potash upon benzaldehyde, so
we must conclude that opianic acid contains an aldehydic
group ; and from the constitution of hemipinic acid it is obvious
that this aldehyde radical must be next the carboxyl group of
opianic acid.

V. The final proof of the presence of an aldehyde group in
opianic acid is furnished by the behaviour of its sodium salt
when distilled with soda-lime.4 Carbon dioxide is split off in
the usual way, and the methyl ether of vanillin is left. The
formula of opianic acid must therefore be that which is shown
below —

1 Beckett and Wright, Trans. Chem. Soc., 1875, 28, 583.

2 Matthiessen and Foster, Annalen Suppl, I. 333; II. 378 ; V. 333.

3 Ibid., I. 332; 11.381.

4 Beckett and Wright, Trans. Chem. 8oc.t 1875, 28, 583.

THE ALKALOIDS 147

OCH3

CHaO— f \_COOH

\— CHO k I1— CHO

Opianic acid. Methyl ether of vanillin.

5. The Constitution of Cotarnine.

The next stage in the proof of the narcotine constitution is
reached through the constitution of cotarnine. This substance l
is obtained along with opianic acid when narcotine is treated
with oxidizing agents —

C22H2307N + 0 + H20 = Ci0H1005 + C12H1504N
Narcotine. Opianic acid. Cotarnine.

I. Cotarnine reacts with two molecules of methyl iodide,
thus proving that it is a secondary base. The reaction
product is called cotarnomethine methyl iodide,2 and has the
composition CiiHnO<iN(CH3)3I.

II. By heating this body with caustic soda, trimethylamine
is split off,3 and cotarnone, CnH1004, remains. This proves to be
an aldehyde, so that its formula can be written Ci0H903 . CHO.

III. When cotarnone is oxidized with potassium perman-
ganate 4 it gives a lactone, cotarnolactone, CnH1006, from which,
on further oxidation, cotarnic acid, CioH807, is obtained.

IY. By the usual reactions it is found that cotarnic acid 5 is
dibasic, contains a methoxyl radical, and has its carboxyl groups
in the ortho-position to one another, as is shown by the ease
with which it forms an anhydride. When heated with
phosphorus and hydriodic acid to about 160° C. it yields gallic
acid —

Wohler, Annalen, 1844, 50, 1.

Boser, Annalen, 1888, 249, 157.

Ibid., 141.

Hid., 163.

Ibid., 163; 1899, 254,341.

148 RECENT ADVANCES /Af ORGANIC CHEMISTRY

HO-I LCOOH

V. Now, gallic acid differs from cotarnic acid by the
group C3H202 —

C10H807 - C7H605 = CH2 + C0
Cotarnic Gallic From methoxy
acid. acid. group.

Part of this we can account for by the loss of carbon dioxide
from a carboxyl group, since cotarnic acid is dibasic, while
gallic acid is monobasic. We have thus one carbon atom left
unaccounted for. This must be derived from the methylene
group of a methylene ether. We are in this way led to
formulate cotarnic acid as a methyl-methylene-gallic-carboxylic
acid, C6H(OCH3)(CH202)(COOH)2. For such a substance there
are only two possible formulae —

CH2— 0 OCH

Without going into details,1 we may say that the general
behaviour of the substance is best represented by (II.). Cotarnic
acid therefore has the constitution —

OCH,

COOH

)— W ]»— COOH
Cotarnic acid.
1 Freuiid and Becker, Her., 1903, 36, 1521.

THE ALKALOIDS 149

VI. Cotarnolactone must therefore have the formula —

OCH3

and cotarnone must be —

OCH3

/o-f VOHO

CH2(

\0— L II— CH=CH2

VII. But cotarnone was obtained from cotarnomethine
methyl iodide and soda, whence cotarnomethine methyl iodide
must have the structure —

OCHg

CH3

CH2

\0— L J— CH2— CH2— N

VIII. Hence cotarnine should have the following constitu-
tion ; since cotarnomethine methyl iodide is obtained from it
by the action of two molecules of methyl iodide —

OCH3

n.
-CHO
-CH2— CH2-NH— CH3

IX. This formula, however, fails to explain the formation
of a pyridine derivative, apophyllenic acid, when cotarnine is

ISO RECENT ADVANCES IN ORGANIC CHEMISTRY

oxidized with nitric acid;1 and to account for this we must
assume that the free aldehydic group has disappeared in the
course of some intramolecular ring-formation, which simul-
taneously brings into existence a pyridine chain within the
molecule of cotarnine. This change we may represent in two
ways, as shown in the formulae below —

OCH3 |
OH

OCH

CHO

c: _

Cotarnine (carbinol form).

Cotarnine (ammonium form).

It is generally agreed that the salts of cotarnine are best
represented as derivatives of the ammonium form ; for instance,
the production of apophyllenic acid can be made clear on this
assumption —

N03 / \ „

Oxidize HOOC— C N/

I \CH3 > || I XCH3

CH2 HOOC— C CH

\ / V/

CH2 CH

Cotarnine nitrate. Apophyllenic acid derivative.

1 Wohler, Annalen, 1844, 50, 24.

THE ALKALOIDS 151

With regard to the free base, however, the spectroscopic
investigations of Dobbie, Lauder and Tinkler l have shown that
the structure varies with the solvent in which the substance is
dissolved. In ether or chloroform the carbinol form is present ;
but the addition of alcohol to the solution brings into existence
the ammonium form; in pure alcoholic solution no less than
25 per cent, of the substance is present as ammonium base.

6. The Synthesis of Cotarnine.

In the last section we dealt with the constitution of cotar-
nine, and we must now take up the synthesis of this substance.
Synthetic cotarnine has been prepared by Sal way ; 2 but as the
constitution of one of his intermediate products is left doubtful
in the synthesis, it is not possible to establish the cotarnine
structure from his work. In the light of the facts given in
the last section, however, we can deduce the formulae of the
intermediate compounds.

I, The first stage in the process is the synthesis of j3-3-
rnethoxy-4 : 5-methylenedioxy-phenyl-propionic acid. Salway
took as his starting-point the substance myristicin —

— ff ^— CH2— CH=CH2

OCH3

which he obtained from oil of nutmeg. This was heated with
alcoholic potash to convert it into iso-myristicin ; and the latter
was then oxidized to myristicin aldehyde by means of potassium
permanganate —

1 Dobbie, Lauder and Tinkler, Trans. Chem. Soc., 1903, 83, 598.

2 Salway, Trans. Chem. Soc., 1910, 97, 1208.

152 RECENT ADVANCES IN ORGANIC CHEMISTRY

0— rf \s-CH-CH— CH

OCH3

Isomyristicin.

OCH3

Myristicin aldehyde.

The aldehyde was then condensed with ethyl acetate by means
of sodium, and the resulting ester was hydrolyzed with alco-
holic potash —

—OHIO BQCH.COOC2H.

OCH,

,0—<f ^— CH=CH— COOC2H5
CH

The substituted cinnamic acid thus produced was reduced with
sodium amalgam, and in this way the required |3-3-methoxy-
4 : 5-methylenedioxy-phenyl-propionic acid was obtained.

THE ALKALOIDS 153

— CH2— CH2— COOH

II, The second stage ends in the production of phenyl-
acetyl-/3- 3- methoxy- 4 : 5-methylenedioxy-phenylethylamine.
The acid (I.) was converted into the amide (II.) in the usual
way, and this in turn was changed into the corresponding
amine (III.) by Hofmann's reaction —

,0— fl NJ— CH2— CH2— CO— NH
CH2

— (f ^— CH2— CH2— NH2

The phenylacetyl derivative (IV.) was then prepared by the
ordinary method —

— CH2— CH2— NH— CO— CH2— C6H5

*3 (IV.)

Phenylacetyl-)8-3-methoxy-4:5-methylenedioxy-phenyl-ethylamine.

III. This phenylacetyl derivative was condensed by heating
it with phosphoric oxide in presence of xylene ; and in this way
a mixture of two isomeric dihydro-isoquinoline derivatives was
produced (V. and VI.).

154 RECENT ADVANCES IN ORGANIC CHEMISTRY

cnX

— CH2— CHa— NH— CO— CH2— C6H5

-^ (IV.)

OCH,

(VI.)

IV. The substance (V.) is 8-methoxy-6 : 7-methylenedioxy-
l-benzyl-3:4-dihydro-isoquinoline. To convert it into cotar-
nine, it is necessary in the first instance to form its metho-
chloride (VII.), which is then reduced by means of tin and
hydrochloric acid to 1-benzyl-hydrocotarnine (VIII.).

CH2

OCH

(VII.)

CH2.C6H5

CH
OCH3|

CH2 . CgHs
(VIII.)

Finally, oxidation with manganese dioxide in presence of sul-
phuric acid converted the benzyl derivative into cotarnine.

^0—

CH2

CH,

NH . CH3
CHO

OCH,

THE ALKALOIDS 155

It will be noticed that the substance (VI.), if treated in the
same way as (V.), would give rise to an iso-cotarnine ; and if
the cotarnine constitution were unknown, this synthesis would
throw no light upon the relative positions of the methoxy-group
and the pyridine ring.

7. The Synthesis of Hydrocotarnine.

On reduction, cotarnine is converted into hydrocotarnine,1
which is formed in the manner indicated by the formulae below —

OCH

CHO

CH2

Alcohol.

OCHg

CH<

^\ ,

,0-1

CH2

x /
CH2

Hydrocotarnine.

1 Beckett and Wright, Trans. Chem. Soc., 1875, 28, 577; Bandow and
Wolffenstein, Per., 1898, 31, 1577.

156 RECENT ADVANCES IN ORGANIC CHEMISTRY

8. The Constitution of Narcotine.

We have now in the course of the previous sections amassed
the material which we require in our consideration of the
narcotine formula ; and we may next proceed to deal with the
question.

Narcotine contains no carboxyl or hydroxyl radical. It is
made up of one hydrocotarnine nucleus and one opianic acid
nucleus, the latter being in the form of the lac tone, meconine.
This is shown by the action of reducing agents upon narcotine —

H2 = C10H1004
Narcotine. Meconine. Hydrocotarnine.

We must now consider the mode of linkage of these two
nuclei. When we examine the formulae of meconine and
hydrocotarnine —

OCH3

I CH2

N.CH
CH2

CH2

Meconine. Hydrocotarnine.

it is obvious that the linking does not take place through
an oxygen atom, as all of these are fully occupied. It must,
therefore, occur by the junction of two carbon atoms, each
of which loses a hydrogen atom in the union. The pair of
atoms which are most likely to be concerned in the linkage
are those which give rise to the aldehyde groups of opianic
acid and cotarnine, so that the formula of narcotine would be
written —

THE ALKALOIDS 157

OCH3

CH30—

N.CH3

CH2

Narcotine.

9. The Syntheses of Gnoscopine and Narcotine.

Perkin and Robinson l showed that when cotarnine and me-
conine are boiled in alcoholic solution in presence of potassium
carbonate the substance produced is identical with the alka-
loid gnoscopine ; and by fractionally crystallizing the d-bromo-
camphorsulphonate of the base 2 they were able to isolate the
dextro and fcevo forms of narcotine, gnoscopine being the racemic
variety. The laevo-narcotine thus obtained was identical with
the natural alkaloid.

10. The Synthesis of Narceme.

When the methyl iodide addition product of narcotine is
treated with alkalis, it is converted into a substance narcei'ne,
which was first called pseudo-narceine.3 The course of the
reaction may be formulated in the following way : —

1 Perkin and Robinson, Proc. Chem. 800., 1910, 26, 46.

2 Perkin and Robinson, ibid., 131.

3 Roser, Annalen, 1888, 247, 167; 1889, 254, 357; Freund and Frankforter,
ibid., 1893; 277, 31.

158 RECENT ADVANCES IN ORGANIC CHEMISTRY

OCH3 OCH3

I
CH30-,/\-CO

—OH/

CH30

KOH

CH.O

0—

N— CH3

1
CH

,0—

CHS

N(CH3)2
CH2

CH2

Narcotine methyl iodide.

CH2

Intermediate product.

OCH

CH30

^0— ^ A CH

CH2

Intermediate product.

CH2— CH2— N(CH3)2

Narceine.

11. The Synthesis of Hydrastinine.

This substance, which occurs among the decomposition
products of the alkaloid hydrastine, has been synthesized by

THE ALKALOIDS 159

Fritsch l ; and as a knowledge of its constitution may help us
in our consideration of the hydrastine formula, we may give
a brief account of Fritsch' s work before dealing with the
natural alkaloid.

When chloracetal is treated with ammonia, it yields the
substance acetalamine, which has the formula —

KH2.CH2.CH(OC2H5)2

This substance can be made to condense with aromatic
aldehydes; and when the products thus obtained are treated
with sulphuric acid, alcohol is split off and isoquinoline
derivatives are formed. If we apply this reaction to the case
of piperonal, we shall have the following series of reactions : —

,0—

CHO + H2N . CH2 . CH(OC2H5)2

V-L .

Acetalamine.
Piperonal.

Q-f ^-CHtN.CHa.CHCOCaHs), + H2O

= CH/

\0— «

Piperonalacetalamine.
CH

/O— (f ^X N -2C2H5OH \

[/ I - — ^ CH2(

\o_il JH CH2 x<

)H(OC2H5)2

Piperonalacetalamine. Methylenedihydroxyisoquinoline.

When the methyl iodide addition product of this body is
reduced by means of tin and hydrochloric acid, it gives the
substance hydrohydrastinine —

1 Fritsch, Annalen, 1895, 286, 18.

60 RECENT ADVANCES IN ORGANIC CHEMISTRY

\0- CH

^
CH

lodomethylate.

CH2

Hydrohydrastinme.

This last substance Freund1 has converted into hydrasti-
nine by oxidizing it with potassium bichromate and sulphuric
acid.

Now, from the fact that the behaviour of hydrastinine, on
reduction and salt formation, closely resembles that of cotarnine,
we are enabled to put forward the following structural formula
for it : —

CHO

NH.CH3
CH2

CH2

Hydrastinine.

CH2

Hydrastinine hydrochloride.

This formula explains why hydrastinine behaves as an alde-
hyde, why it forms a ring compound in presence of acids, why
its salts contain one molecule of water less than the free base,
why it yields apophyllenic acid on oxidation, and many other
properties which the substance possesses. A comparison of
their formulae will show that cotarnine is a methoxylated
hydrastinine.

12. The Constitution of Hydrastine.

Hydrastine contains one methoxyl group less than narco-
tine, but in all other respects it resembles that compound.
Now, on oxidation with dilute nitric acid, hydrastine breaks
down into hydrastinine and opianic acid just as narcotine
breaks down into cotarnine and opianic acid. But, as was
shown in the preceding section, cotarnine is methoxy-hydrasti-
nine, so that we may conclude that if we eliminate the

1 Freund, Ber., 1887, 20, 2403.

THE ALKALOIDS 161

inethoxy-group from narcotine we shall have hydrastine. This
actually proves to be the case ; so that we may write the
formula of hydrastine by simply taking that of narcotine and
replacing the methoxyl radical of the cotarnine half by a
hydrogen atom. Hydrastine would therefore be —

OCHg

CH30-(f ^)-

Hydrastine.

G. — THE PURINE* GROUP.
1. The Synthesis of Uric Acid.

The problem of the constitutions of the purine derivatives
has proved one of the most complicated chapters in the recent
history of organic chemistry; so complicated is it that we
cannot devote sufficient space to allow of any historical treat-
ment of the matter, but must confine ourselves as closely as
possible to the actual proofs of the constitutions of some of the
purine series.

The most important member of the group is uric acid. This
substance l has been synthesized in a variety of ways ; but for
the most part the syntheses throw no very clear light upon the

* This, like many other chemical terms, is what Lewis Carroll defined as a
portmanteau word ; it is derived from the two words purum uricum.

1 Horbaczewski, Monatsh., 1882, 3, 796 ; 1885, 6, 356 ; 1887, 8, 201, 584 ;
Behrend and Roosen, Ber., 1888, 21, 999 ; Annalen, 1889, 251, 285 ; Traube,
Ber., 1900, 33, 1371, 3035 ; Fischer and Ach, Ber., 1895, 28, 2473; Fischer, Ber.,
1897, 30, 559.

162 RECENT ADVANCES IN ORGANIC CHEMISTRY

constitution of the body. We may describe very briefly two of
these synthetic methods of preparing uric acid, the first being
due to Emil Fischer and the second to W. Traube.

When malonic acid is treated with urea, it yields a cyclic
ureide. malonyl-urea or barbituric acid —

NH2 HO— CO NH— CO

I I I

CO CH2 = 2H20 + CO CH2

II II
NH2 HO— CO NH— CO

Barbituric acid.

If we treat barbituric acid with nitrous acid, the methylene
group is replaced by the isonitroso-radical in the usual way,
giving us oximido-malonyl-urea, which is also called violuric
acid ; and on reduction of this substance the oximido-group is
converted into an amido-radical, producing amido-malonyl urea,
or uramil—

NH— CO NH— CO NH— CO

I I II

CO CH2 CO C:NOH CO CH.NH2

-CO NH— CO NH— CO

Barbituric acid. Violuric acid, Uramil.

On treatment with potassium cyanate, uramil takes up cyanic
acid and is changed into pseudo-uric acid —

NH— CO NH— CO

II II

CO CH.NH2 CO CH.NH.CO.NH2

NH-CO NH-CO

Uramil. Pseudo-uric acid.

It is very hard to extract water from pseudo-uric acid, but
this can be done by heating it with molten oxalic acid or by
boiling it with hydrochloric acid. Under these circumstances
one molecule of water is lost and uric acid is formed. Uric
acid should therefore have the following constitution : —

THE ALKALOIDS 163

NH— CO

CO CH— NH\

I I /CO

NH— C=N^

Its property of forming salts could be ascribed to the
existence of an enolic form, such as —

N-C.OH

. C C - N^

II II

N— C -- X

It is more usual, however, to consider uric acid to exist in
the isomeric form —

NH— CO

CO C-NHX

I ll ;co

NH— C— NH/

Uric acid.

The second synthesis takes as its starting-point the conden-
sation of urea with cyanacetic acid, which takes place under the
influence of phosphorus oxychloride —

NH2 HO— C:0 NH— CO

CO CH2 CO CH2

NH2 CN NH2 CN

Cyanacetyl-urea.

Caustic soda causes cyanacetyl-urea to undergo an intra-
molecular change by which it is converted into amido-uracil —

NH— CO

CO CH

I II
NH— C— NH2

Aimdo-uracil.

1 64 RECENT ADVANCES IN ORGANIC CHEMISTRY

When this is treated with nitrous acid it gives a nitroso-
compound which can be reduced with ammonium sulphide to
diamido-uracil —

NH— CO NH— CO

CO C.NO CO C.NH2

I! I II

NH— C . NH2 NH— C . NH2

Nitroso-eompound. Diamido-uracil.

The next step is to treat this diamido-derivative with caustic
potash and chloroformic ester, by which means a urethane is
formed —

NH— CO NH— CO

I | Cl.COOEt I
CO C.NH2 > CO C.NH.COOEt

I II I II

NH— C.NH2 NH— C.NH2

Diamido-uracil. Diamido-uracil urethane.

By heating the sodium salt of this substance to 180°-190° C.
we obtain the sodium salt of uric acid.

By adapting this last synthesis we can obtain many uric
acid derivatives ; for we may use substituted ureas instead of
the parent substance, or we may replace the urea by guanidine,
or, la&tly, we may discard the chloroformic ester in favour of
formic ester.

Before leaving the question of uric acid we must glance for
a moment at the behaviour of that substance when treated with
various oxidizing agents.

"When the oxidation is carried out by means of cold nitric
acid, the six-membered ring of uric acid remains intact, while
urea is split off. The oxidized ring which remains can be
derived from mesoxalic acid and urea ; it is termed alloxan, or
mesoxalyl-urea —

NH— CO NH— CO

CO C— NHV CO CO H2NX

;co N;co

NH— C(

NH— C— NHX NH— CO H2N

Uric acid. Alloxan. Urea.

THE ALKALOIDS 165

If, on the other hand, we use alkaline potassium perman-
ganate solution as our oxidizing agent, the five-membered ring
remains unbroken, while the six-membered one is destroyed.
The first products in this case are two substances, uroxanic
acid, CsHsN^e, and oxonic acid, C^HsNaO*, which are further
oxidized to allantoin —

NH2

CO CO— NHX

I I >o

NH— CH— NHX

Allantoin.

With hydrogen peroxide the sodium salt of uric acid yields
a substance of the formula C^HaN^O^ tetracarbonimide, which
acts as a weak tetra-basic acid ; on this account the following
formula has been tentatively ascribed to it : —

NH— CO— NH

I I

CO CO

NH— CO— NH

2. The Synthesis of Theophylline.

If in the uric acid syntheses we substitute symmetrical
dimethyl-urea for the parent substance, we obtain in the end
dimethyl uric acid —

CH3— N CO

CO C— NHX
I II )CO
CH3— N C— NHX

When this is treated with trichloride and oxy chloride of
phosphorus at 150° C. it is converted into a substance chloro-
theophyllin, one atom of chlorine replacing a hydroxyl group.
Chlorotheophylline must, therefore, have the following con-
stitution : —

1 66 RECENT ADVANCES IN ORGANIC CHEMISTRY

CH.q— N CO

CO C— NH

I II
CH3— N C— N^

XC.C1

By reducing with hydriodic acid, theophylline l is formed —

CH3— N -- CO

CO C— NIL

I II
CH3— N -- C— N

Theophylline.

3. The Synthesis of Caffeine.

Caffeine l is obtained by the action of methyl iodide upon
theophylline. Its constitution is therefore expressed by —

CH3— N -- CO

CO C— Nr-CH3

II )CH
CH3— N -- C— 1SK
Caffeine.

4. The Synthesis of Theobromine.

If we take as a starting-point the dimethyl uric acid which
has the constitution (I.) shown below, and treat it with
phosphorus oxychloride, we shall find that it gives chlorotheo-
bromine (II.), which, on reduction with hydriodic acid, yields
theobromine (III.).2 The reactions are parallel to those which
lead from the isomeric dimethyl uric acid to theophylline.

CO 0_N-CH3 ,0

)co i ii j.ci

CH3-N. _ C-NH CH3-N- -C-N^

(I.) (II.)

1 Fischer and Ach, Her., 1895, 28, 3135.

2 Fischer, Ser., 1807, 30, 1839.

THE ALKALOIDS

167

NH-CO
CO C—

N-CH3

CH3— N -- C—

Theobromine.

(in.)

5. The Synthesis of Purine.

When the sodium salt of uric acid is treated with
oxychloride of phosphorus it yields a hydroxy-dichloro-purine
of the following formula : —

N=0.d

Cl.C C— NH

;c . OH

N— C— N

This, by means of trichloride of phosphorus, can be changed
into a trichloro-derivative, the third hydroxyl group being
replaced by a chlorine atom. The substance thus formed,
trichloropurine, is then treated with hydriodic acid at 0° C.
whereby di-iodopurine is produced. This, by reduction with
water and zinc dust, gives purine itself.

01 . C C— NH

N— C— N
Trichloropurine.

N=C.I

I I
I.C C— NH

)CH

N— C— N
Di-iodopurine.

CH C— NH

N - C— N
Purine.

Purine is the substance to which all the substances of the
purine group are usually referred; the derivatives being dis-
tinguished by means of the system of numbering shown in the
following scheme : —

168 RECENT ADVANCES IN ORGANIC CHEMISTRY

(1) (6)
N 0

I I (7)
(2) C (5) C Nx

I I > (8)

N 0 W

(3) (4) (9)

According to this, the substance xanthine is 2, 6-dihydroxy-
purine ; theophylline would be 1, 3-dimethyl-xanthine ; caffeine
would be 1, 3, 7-triinethyl-xau thine ; theobromine would be
3, 7-dimethyl-xanthine ; and uric acid 8-hydroxy-xauthine.

CHAPTER VII

THE POLYPEPTIDES l

WHEN we examine the contents of the cells from which living
tissues are built up, we find that they are for the most part
made up of albuminous bodies of extremely complicated chemical
character. These albumins are distinguished from all the other
naturally occurring substances by the fact that animal life may
be supported upon them alone in conjunction with water and
salt; whereas fats and carbohydrates do not in themselves
furnish nourishment sufficient for the support of animal functions
for an indefinite period. The importance of the albumins from
the physiological point of view, therefore, can hardly be over-
estimated; while from the chemical side they furnish one of
the most difficult and complicated problems which the organic
chemist has yet attacked.

The difficulties of the researches which have been carried
out in this branch of organic chemistry can hardly be over-
estimated. In the first place, many albumins are non-
crystalline substances which require special treatment before
they can be obtained in crystalline form ; this, of course, makes
it very difficult to determine the state of purity of any specimen
under consideration. Secondly, the extreme sensitiveness of
albumins to heat, acids, or alcohol renders them very liable to
be altered during the progress of the ordinary chemical reactions.
Again, the molecular complication of these substances must be
tremendous, if we are to judge from molecular weight determina-
tions : egg albumin has been estimated to have a molecular

1 A complete set of references up to 1906 will be found in a lecture by
Fischer (Ber., 1906, 39, 530). See also Fischer, Ber., 1906, 39, 2893; 1907, 40
1754, 3704 ; 1908, 41, 850, 2860 ; Fischer and Konigs, Ber., 1907, 40, 2048 ;
Fischer and Schulze, ibid., 943 ; Fischer and Gerngross, Ber., 1909, 42, 1485 ;
Fischer and Luiiiak, ibiiL, 4752. Fischer's papers have been reprinted in his
book " Die Aminosauren, Polypeptide und Proteine" (1906).

i;o RECENT ADVANCES IN ORGANIC CHEMISTRY

weight of at least 15,000, according to the results of the freezing-
point method.

In the foregoing chapters we have dealt at some length
with the constitutions of various compounds, and it will be
remembered that there are two general methods of investigating
the constitution of any given substance. We may attack the
question from the synthetical side or from the analytical point
of view: in the first case we study the general properties
of the substance, then ask ourselves in what way we can
build up a molecule whose reactions will resemble those of
the one we are studying, and having synthesized this body
we compare its reactions with those of the original; in the
analytical method, we take the molecule to pieces in various
ways, and isolate a series of decomposition products, from
which we endeavour to guess the manner in which they were
arranged in the original molecule. Now, in the case of the
albumins, the first line of research turned upon the analytical
results. This was to be foreseen, for it seemed almost im-
possible to build up molecules of such extreme complexity.
The analytical method, however, has not carried us very far ;
and the most important work on the question has been carried
out from the synthetical side since Fischer attacked the problem.
Before dealing with his work, however, we must cast a glance
at the decomposition products which have been obtained from
the albumin group.

The oxidation of the albumins cannot be said to have
yielded results of any great interest; the greater part of our
knowledge of these bodies has been obtained by means of
hydrolysis reactions. When ferments are allowed to act upon
protein derivatives, the bodies first formed are albumoses and
peptones. These intermediate compounds can be further broken
down into amido-acids. Hydrolysis by means of alkali takes
place more rapidly, while acids decompose the albumins most
easily. It is thus made clear that the substances lying at the
base of the albumins belong to the class of amido-acids ; and,
further, that these acid nuclei are linked together in some way
which allows them to be separated one from another by means
of hydrolysis. It is evident that amide-formation is the most
probable method of uniting the nuclei ; and from this point of
view Fischer took up the work of synthesizing some compounds

THE POLYPEPTIDES 171

which, while not themselves of the protein class, would show
sufficient resemblance to the naturally occurring substances to
allow us to deduce the probable constitution of at least part of
the albumin molecule.

To describe these synthetic substances, Fischer proposed the
name " Polypeptides" by which he intends to denote those
compounds which are derived from two amido-acid molecules
by the elimination of water. A few polypeptides have been
obtained by the hydrolysis of proteins, but by far the greater
number are synthetic. We may now give the outlines of the
methods employed by Fischer in his researches.

In the first place, it was necessary to obtain mono-amido-
acids. This Fischer did by means of the ordinary methods —
action of ammonia on the esters of bromo-fatty acids or by
Strecker's cyanhydrin method (addition of hydrocyanic acid and
ammonia to an aldehyde and hydrolysis of the cyanhydrin thus
formed). Now, having obtained these acids, another problem
presents itself. If we combine together two racemic acids we
shall have not a single reaction product, but a mixture of two
new racemic substances. For instance, if we start with racemic
alanine and racemic leucine, we should produce a mixture of
the four isomers—

d-Alanine-^-leucine. c£-Alanine-/-leucine.
£-Alanine-Z-leucine.

The two substances in the left-hand column then combine to
form a racemic substance, and the two in the right-hand column
to form another racemic compound, so that we should have two
new bodies instead of a pure compound. And, of course, if we
coupled together more than two racemic acids we should find
the number of stereo-isomers in the product increased in like
manner. This evidently threw considerable difficulty in the
way, and to avoid it Fischer resolved to use in his condensations
optically active acids only. By this means he excluded the
possibility of racemic compounds being formed, so that from one
pair of amido-acids he obtained only a single reaction product.

This did not clear the experimental difficulties away, how-
ever; it only carried them one step further back. For, owing
to the very weak acidity of the amido-acids, resolution of these
substances into their optically active antipodes by salt-formation

i?2 RECENT ADVANCES IN ORGANIC CHEMISTRY

with active bases was by no means an easy task. Fischer
evaded this difficulty in turn by one of his usual simple artifices.
He benzoylated the amido-group of the acid, and thus reduced
its basic properties to a minimum; thereafter, resolution into
the optical antipodes presented no difficulty, and after this had
been accomplished, the benzoyl radical was split off and the
optically active amido-acid remained.

I. The first method employed by Fischer in the actual
synthesis of polypeptides depends upon the elimination of a
molecule of alcohol from two molecules of amido-acid ester —

. CH2 . COOEt + NH2 . CH2 . COOEt

2 . 2 .
= NH2 . CH2 . CO . NH . CH2 . COOEt + EtOH

Now, it will be seen at once that if we applied this method
as given above to a mixture of two different amido-acid s, it would
be sheer chance that would govern the production of the end-
product. For example, if we were to combine together the two
esters (A) and (B) we should get a mixture of (C) and (D) in the
reaction product —

(A) NH2.CH2. COOEt
* (B) NH2 . CH . COOEt

CH3

(D) NH2.CH. CO. NH.CH2. COOEt

CH3

This difficulty in its turn was overcome by Fischer in a very
simple manner. Before condensing the two substances together
he allowed one of them to react with ethyl chlorocarbonate,
which acted upon the amido-group and protected it from
further attack—

Cl . COOEt + NH2 . CH2 . COOEt

= EtOOC . NH . CH2 . COOEt + HC1

When a compound such as this is heated for thirty-six hours
with the ester of an amido-acid, alcohol is eliminated between
the — NH2 group of the amido-acid and the — CH2 . COOEt group

THE POLYPEPTIDES 173

of the above substance, whose amido-group cannot react in this
way. Thus we know at once the constitution of the resulting
compound. An example will serve to make the matter clear.
If we start with the substance glycyl-glycine,* and treat it with
chloro-carbonic ester, we shall obtain the substance shown below,
glycyl-glycine carboxylic acid ester —

EtOOC . 01 + NH2 . CH2 . CO . NH . CH2 . COOEt

= EtOOC . NH . CH2 . CO . NH . CH2 . COOEt + HC1

When this substance is heated for thirty-six hours with
leucine ester, ethyl alcohol is eliminated in the following
way: —

EtOOC.NH.CH2.CO.NH.CH2.COOEt+NH2.CH.(C4H9).COOEt
= EtOOC.NH.CH2.CO.NH.CH2.CO.NH.CH(C4H9).COOEt

This substance is the carboxylic ester of glycyl-glycine-
leucine ; as can be seen from the formulae, it can have no other
constitution than that shown. This carbethoxy-glycylglycyl-
leucine ester contains three amido-acid nuclei, and is therefore
called a tri-peptide derivative.

II. The yields of end-product from the foregoing method of
synthesis were poor, and Fischer therefore turned to another way
of attaining his objective. When the ester of the chlorocarbonic
derivative of an amido-acid is treated with thionyl chloride, an
acid chloride is formed ; and this readily condenses with amido-
acid esters, forming polypeptide derivatives. For instance, if we
start again with the derivative obtained by the action of chloro-
carbonic ester upon glycylglycine, and treat it with thionyl
chloride, we shall produce the chloride whose constitution is
shown below —

EtOOC . NH . CH2 . CO . NH . CH2 . CO . Cl
When this chloride is condensed with glycylglycine ester —
NH2 . CH2 . CO . NH . CH2 . COOEt

it yields the tetra-peptide derivative, glycylglycylglycylglycine-
carbethoxy-ester —

EtOOC.NH.CH2.CO.NH.CH2.CO.NH.CH2.CO.NH.CH2.COOEt

* Fischer terms "glycyl" the radical NH2.CH2.CO— which is derived
from glycine (glycocoll) NH2 . CH2 . COOH,

174 RECENT ADVANCES IN ORGANIC CHEMISTRY

III. The drawback of the two foregoing methods lies in
the fact that so far no method has been discovered by
means of which we can eliminate the group — COOEt, which
is attached to one end of the polypeptide chain; so that
neither method can be employed to build up a true polypeptide.
Fischer therefore devised another method by means of which
the polypeptides themselves can be produced. Starting from
the ester of a substance like glycine (I.) or glycylglycine, he
treated this with chloracetyl chloride (II.) or some similar com-
pound. Hydrochloric acid is eliminated, and the two molecules
combine together to form a compound with chlorine at one end
of the chain (III.)- The ester group at the other end of the
chain is then hydrolyzed very carefully, and a chloro-acid
produced (IV.), which, on treatment with ammonia, yields a true
polypeptide (V.) —

(I.) NH2.CH2. COOEt

(II.) C1.CH2.CO.C1
(III.) 01 . CH2 . CO . NH . CH2.COOEt
(IV.) 01 . CH2 . CO . NH . CH2 . COOH

(V.) NH2 . CH2 . CO . NH . CH2 . COOH

The reason for hydrolyzing the ester (III.) to the acid (IV.)
lies in the fact that if this were not done an amide would be
formed on treatment with ammonia, and the amido group would
be most difficult to get rid of later.

IV. A variation of the previous method may also be used.
If we take the substance —

Cl . CH2 . CO . NH . CH2 . COOH

which was formed in the course of the last synthesis we
described, and treat it with pentachloride of phosphorus, we
convert the acid into the chloride * —

Cl . CH2 . CO . NH . CH2 . CO . Cl

which can then be made to interact with glycine ester, yielding
the more complicated substance —

Cl . CH2 . CO . NH . CH2 . CO . NH . CH2 . COOEt

* Thionyl chloride is a better reagent than phosphorus pentachloride for
producing acid chlorides. The reaction takes place according to the equation —

K . COOH + SOC12 = K . CO . Cl + S02 + HC1

from which it will be clear that the acid chloride can be obtained pure simply
by boiling off the sulphur dioxide and hydrochloric acid.

THE POLYPEPTIDES 175

The remaining chlorine atom may then be replaced by the
amido-group by means of ammonia ; and after hydrolysis of the
ester group the tri-peptide glycylglycylglycine is formed —

NH2 . CH2 . CO . NH . CH2 . CO . NH . CH2 . COOH

V. This modification has been farther extended. When
amido-acids are treated with a mixture of acetyl chloride and
phosphorus pentachloride, the corresponding acid chlorides are
formed. These can be combined with other amido-acids, and in
this way we can obtain polypeptides. For instance, if we take
glycine and treat it as described we should expect to produce
glycyl chloride —

NH2.CH2.COOH -> NH2 . CH2 . CO . Cl

This can be condensed with another molecule of glycine,
forming glycylglycine —

NH2 . CH2 . CO . 01 + NH2 . CH2 . COOH

= NH2 . CH2 . CO . NH . CH2 . COOH

VI. If we abstract two molecules of alcohol from two mole-
cules of an a-amido-ester, a cyclic substance is produced, which
is a derivative of ay-diketo-piperazine —

CH2— NH2 EtO— CO CH2— NH— CO

+ = | + 2EtOH

CO— OEt H2N— CH2 CO — NH— CH2

This cyclic compound, when carefully treated with hydro-
chloric acid, can be opened out into an open-chain body,
glycylglycine —

CH2— NH— CO CH2— NH2 COOH

I | +H20= | |

CO — NH— CH2 CO — NH— CH2

By choosing the appropriate amido-ester from which to start,
a given polypeptide may be obtained in this manner.

We cannot go into details with regard to the various
substances which have been synthesized by means of the
foregoing methods, but there is one substance which is worthy
of mention. Fischer has recently synthesized an octadeca-
peptide in the following manner. Starting from dextro-
a-bromo-isocapronyl-diglycylglycine —

176 RECENT ADVANCES IN ORGANIC CHEMISTRY
Br . CH . CO . (NH . CH2 . C0)2 . NH . CH2 . COOH

he treated this according to the fourth method, combining it
with pen ta-glycylgly cine, and, finally, exchanging the bromine
atom for an amido-group, he obtained Isevo-leucyl-octaglycyl-
glycine —

NH2 . CH . CO . (NH . CH2 . C0)8 . NH . CH2 . COOH

This polypeptide was then coupled with dextro-bromo-iso-
caproyl-diglycylglycine, and again treated with ammonia,
whereby the tetradecapeptide shown below was formed —

NH2.CH.CO.(NH.CH2.CO)3.NH.CH.CO.(NH.CH2.CO)8.NH.CH2.COOH

C4H9
Lasvo-leucyl-triglycyl-laevo-lucyl-octaglycyl-glycine.

C4H9

By a repetition of this series of reactions the octadecapep-
tide was formed, which has the constitution shown below —

NH2.CH(C4H9).CO.(NH.CH2.CO)3.NH.CH(C4H9).CO.(NH.CH2.CO)3.NH

HOOC.CH2.NH.(CO.CH2.NH)8.CO.CH.C4H9

Z-Leucyl-triglycyl-Meucyl-triglycyl-Meucyl-octaglycyl-glycine.

This extraordinary substance is apparently one of the most
complicated systems of known constitution which has hitherto
been synthesized. Its molecular weight is twelve hundred and
thirteen; while that of the fairly complicated natural body,
tri-stearin, is only eight hundred and ninety-one.

We must now briefly summarize the main characteristics of
the polypeptide class, and it may be of interest to compare them
with those of the naturally occurring proteins. The polypep-
tides are solids, which usually melt at about 200° C., with some
decomposition. They are easily soluble in water, but insoluble
in alcohol, like some of the albumins ; and instead of having
the usual insipid or sweet taste of the ordinary amido-acid,
they are bitter, like the protein derivatives. In dilute sulphuric
acid solution they are precipitated by phosphotungstic acid,
in which behaviour they resemble the albumins. Both the
natural and artificial classes give the biuret reaction. The

THE POLYPEPTIDES 177

action of ferments, or of acids or alkalis, is the same in both
classes; and similar products are obtained when animals are
fed with polypeptides and albumins. In the case of ferment
action it is found that much depends upon the groups which
have been used in building up the polypeptide structure, some
polypeptides being much more easily fermented than others.

From these data it will be obvious that the researches of
Fischer and Curtius have carried us into a series of substances
which, in many respects, resemble the natural bodies of the
protein class ; how far the parallel holds good, and how high in
the scale we can carry our syntheses remains for the future to
make clear.

CHAPTEE VIII

% THE POLYKETIDES AND THEIR DERIVATIVES

BROADLY speaking, plants differ from animals in that they can
nourish themselves with water and carbon dioxide alone, while
"the animal kingdom requires the intermediation of vegetables
and other organized matter. The substances which lie at the
base of all syntheses of organized tissues must therefore be
simple compounds of carbon, hydrogen, and oxygen. Once
having synthesized such substances, the plant, as will be shown
later in this chapter, could easily build up derivatives of the
aliphatic, aromatic, and heterocyclic series.

Given formaldehyde, sugars may be produced by the action
of alkalis ; and many such examples of the production of com-
plicated natural bodies from very simple substances are known.
In the present chapter we shall confine ourselves to deriva-
tives of one class ; but as this class is interwoven with all the
main groups of organic compounds, it will serve as a skeleton
from which the relations between apparently quite dissimilar
groups can be deduced. At the same time it must be borne in
mind that our laboratory synthetic methods differ in the main
from those employed in the living plant. While we, in our
syntheses, start from the same elements as the plant does, we
usually build up our substances step by step, proceeding from
simple to complex. The plant appears to act differently ; for
it, apparently by condensation, polymerization, or some such
process, converts its simple starting substance into an extremely
complicated derivative, which then decomposes, yielding those
products which have been identified in saps and tissues. Again,
while most of our ordinary laboratory reactions can be applied
to the production of substances which are found in plants, it is
obvious that the plant must obtain the same result in a much
simpler manner. For instance, when we wish to attach side-
chains to a benzene nucleus, we employ aluminium chloride in

THE POLYKETIDES AND THEIR DERIVATIVES 179

the Friedel-Crafts reaction ; but such a reagent could not exist
in a plant. Further, a great number of our laboratory reactions
require the use of high temperatures, which would be fatal to
plant-life.

When we examine the compounds known to us in the
domain of organic chemistry, it is inconvenient for our present
purpose to regard them from the point of view of text-book
classification. What is of chief importance to us is the ques-
tion, Can they be made to react easily ? From this point of
view we divide compounds at once into two groups, the satu-
rated and the unsaturated, the latter being the reactive ones.
This is, of course, speaking in very general terms, for many
saturated substances are quite reactive. Now, among un-
saturated substances we can again distinguish two classes — the
desmotropic and the non-desmotropic. Of these, the desmo-
tropic class is by far the most reactive. The cause of this is
obvious, for if a non-desmotropic substance be brought into the
presence of a reagent, it has only one way in which it can react,
and if combination does not take place the matter ends ; with
a desmotropic compound, however, if the first form fails to
attack the reagent, there is always the possibility that the
second form may be more successful.

In nature, we find many desmotropic and tautomeric sub-
stances ; but the preponderating class is that which contains
compounds of the type

E— CO— CH2— K' -> K— C(OH) : CH— K'

This "keto-enol" type is very widely distributed among
naturally occurring substances ; it is found in nearly every
important class of compounds, from the purine group to the
terpenes ; and, further, its one form is converted into the other
isomeride more easily than is the case with practically any
other mode of isomeric change.

The simplest member of the class of substances containing
this atomic grouping is the compound keten, which has the
formula —

CH2 : CO

As will be seen later, this group of five atoms is capable of
polymerizing or condensing with other compounds in many

i8o RECENT ADVANCES IN ORGANIC CHEMISTRY

ways; and as some general name is required for the whole
series, we shall adopt the proposal of Collie,1 and designate as
" Polyketides " those substances which are obtained by the poly-
merization of keten, and subsequent addition of other atoms.
For example, acetic acid, H . CH2 . CO . OH, which is derived
from keten by the addition of water, would be termed a " mono-
ketide "; while acetoacetic acid, H . CH2 . CO . CH2 . CO . OH,
would be a " di-Jcetide."

The rest of this chapter will be devoted to the discussion
of these classes of compounds. In the first place, a rapid
survey of the general relations of the group to the rest of
organic compounds will be given, after which the reactions and
properties of individual polyketide derivatives will be dealt
with in so far as they concern the main principles of the
subject.

When keten is allowed to stand under pressure at ordinary
temperatures, it becomes converted into the dimolecular form,
diketen, whose constitution will be discussed later. On treat-
ment with pyridine in benzene solution, this diketen (or keten
itself) can be transformed into dehydracetic acid.

There is some dispute as to the actual formula of the last
body, but for the present we may adopt one of those proposed,
and deal with the whole question later.

CH3.CO.CH2.C CO

II I

CH CH2

Dehydracetic acid.

This substance, dehydracetic acid, is one of the most
important of the polyketide derivatives, at least from the
synthetic point of view. From it, by three different reactions,
we can prepare derivatives of the benzene series, the pyridines
and the pyrones. The first of these is obtained by the action
of alkalis on dehydracetic acid, and the reaction probably takes
the following course : —

1 Collie, Proc. Cltem. Soc., 1907, 23, 230.

o
o

Q
?i

W
o

— o

W
o

o*
o

o
o

1 82 RECENT ADVANCES IN ORGANIC CHEMISTRY

In this way both dihydroxyphenyl-acetic acid and orcinol
are formed.

The two other reactions mentioned, resulting in pyridine
and pyrone derivatives, depend upon the intermediate formation
of diacetylacetone from dehydracetic acid in the following
manner : —

0
CH3.CO.CH2.C CO

-C02 CHo.CO.CHo.CO

As soon as the diacetylacetone is formed, it may lose water and
form dimethyl-pyrone, thus —

OH HO 0

CH3.C C.CH3 -H20 CH3.C C.CH3

II II > II II

H.C C.H H.C C.H

Diacetylacetone. Dimethyl pyrone.

or it may, in presence of ammonia, lose two molecules of water
and take up instead the imino-group in the following way : —

THE POLYKETIDES AND THEIR DERIVATIVES 183
NH

H H NH

OH HO / \

/ \ CH3.C C.CH3

CH3.C C.CH3 -2H20 || ||

II || > H.C C.H

H.C^ /C.H V /

- CO^ CO

Diacetylacetone. Lutidone.

These reactions by no means exhaust the possibilities of
diacetylacetone, however, for from it we may produce benzene,
naphthalene, or isoquinoline derivatives by the following steps.
In the first place, by loss of one molecule of water, we can
produce orcinol —

yH3

CH3

O^f

HfJ

CH2 -H.,O

H.(A

"*T T

-•fV

jo '

iox

1 II

HO.C^ C.OH

\)H

Orcinol

This reaction takes place in strong alkaline solutions, but if
we make the alkali very dilute we can extract two water mole-
cules from diacetylacetone in a different way, two diacetylace-
tone molecules being involved : —

CH3. CO . CH : C . CH2. CO . CH3

H2 OH

H . C . CO . CH2 . CO . CH2CO . CH3

CH3.C. CH:C.CH2.CO.CH3

-2H20 || |

> H . C— CO— CH . CO . CH2 . CO . CH3

CHg.C C.CH2.CO.CH3

> II I

H.C C.CO.CH2.CO.CH3

1 84 RECENT ADVANCES IN ORGANIC CHEMISTRY

This new benzenoid compound in turn is susceptible to the
action of less dilute alkalis, losing another molecule of water
and forming the naphthalene derivative shown below —

CH CH2

CH3.CX C CO— CH3

I II

H.C C CHsr-CO.CH8

C CO

CH CH2
-H2o CH3.C C C.CH3

H.C C C . CO . CHc

\ /\ /
C CO

CH3.
H.C

CH CH
\/ \

C C.CHg

II I

C C . CO . CH3

/\ ^

C C

THE POLYKETIDES AND THEIR DERIVATIVES 185

By allowing the benzene compound to stand at ordinary tem-
peratures in presence of ammonia, an isoquinoline derivative
is obtained —

CH CH

CH3.C C C(OH)— CH3

II I

H.C C

CX C(OH)

I II

OH CH— CO— CH3

CH CH

/ \/ \
CH3.C C C.CH3

III
H.C C NH

\//\/

C C

I II

OH CH . CO . CH3

CH CH

CHg.C C C.CHg

> II I I

H.C C N

\ </ \ #
C C

OH CH2COCH3

We may summarize these reactions in the following scheme,
and then proceed to the examination in detail of the various
substances to which we have called attention.

i86 RECENT ADVANCES IN ORGANIC CHEMISTRY

Keten --- > Acetic acid

Polymerization
V

Acetyl-keten — — > Acetoacetic acid
Triacetic lactone ! Polymerization

*"""""'- Dehydracetic acid — > Isomeric acid

Heating

Alkali Acids ^ acids

Orcinol <- ^-Orcinol Pimethyl-pyrone Lutidone

JBa(OH)2
Diacetylacetone

I Very dilute alkali
V
Benzene derivative

Dilute alkali
4-

Naphthalene derivative Isoquinoline derivative.

The root-substance of the polyketide class, keten itself,
was first prepared 1 by the action of a white-hot platinum wire
upon acetic anhydride. About a year later it was shown '2 that
it could also be obtained by the removal of the bromine from
bromacetyl-bromide, a method which has been applied in the
case of ketene homologues also. Keten itself is a colourless
gas at ordinary temperatures, condenses in solid carbon dioxide
to a colourless liquid, and rapidly polymerizes to a brown oil.
Both keten and its polymers have a peculiar penetrating odour.

Keten may be considered to be a second anhydride of acetic
acid ; for, just as ordinary acetic anhydride is obtained by the
removal of a molecule of water from two molecules of acetic
acid, keten is obtained by withdrawing a molecule of water
from a single molecule of the acid.

CH3.CO.OH CH3.CO\

- H20 = )0

CH3.CO.OH CH3.COX

H OH

| | - H20 = CH2:CO

CH2 . CO

1 Wilsmore and Stewart, Nature, 1907, 75, 510 ; Wilsmore, Trans. Chem. Soc.,
1907, 91, 1938.

2 Staudinger and Klever, Ber., 1908, 41, 594.

THE POLYKETIDES AND THEIR DERIVATIVES 187

From this it follows that keten should give most of the usual
anhydride reactions, and this has been shown to be the case.
When passed into water it yields acetic acid \ alcohol reacts
with it to form ethyl acetate; aniline produces acetanilide;
and ammonia forms acetamide. Keten interacts with bromine
to form bromacetylbromide, with hydrochloric acid to give
acetyl chloride. Thioacetic anhydride is produced by the action
of liquid sulphuretted hydrogen upon keten.1

"We have already mentioned that if keten is allowed to
stand at ordinary temperatures it yields a brown condensation
product. When this is distilled, the dimolecular polymer of
keten passes over as a clear liquid having a very pungent
odour.2 This dimolecular keten, like its parent substance, is a
very reactive body. When added to water, it slowly dissolves,
yielding a strongly acid solution which, on boiling (especially
in presence of hydrochloric acid), gives up acetone and carbon
dioxide, losing its acid properties in the process. When it is
added to aniline, dimolecular keten forms acetoacetic anilide,
while with phenylhydrazine it produces a hydrazone-hydrazide.
These reactions are easily explained by assuming that the
substance is acetyl-keten, for then the three cases mentioned
may be expressed by the following equations : —

CH3 . CO . CH : CO + H20 = CH3 . CO . CH2 . COOH

= CH3.CO.CH3 + C02

CH3 . CO . CH : CO + NH2 . C6H5 = CH3 . CO.CH2 . CO.NH.C6H5
CH3 . CO . CH : CO + 2NH2 . NH . C6H5

On the other hand, Staudinger 3 put forward the view that
diketen was a tetramethylene derivative, and Wilsmore,4 after
a thorough examination of the compound's properties, inclines
to this view. He finds that when diketen, dissolved in carbon
disulphide or carbon tetrachloride, is heated first with bromine
and then with alcohol, it yields y-bromo-acetoacetic ester.
This tends to prove that diketen has the structure —

1 Chick and Wilsmore, Proc. Chem. 800., 1908, 24, 78.

2 Ibid., Trans. Chem. Soc., 1908, 93, 946.

3 Staudinger, Ber., 1909, 42, 4908.

4 Wilsmore, private communication to the author.

iSS RECENT ADVANCES IN ORGANIC CHEMISTRY

CHo— CO

for on this assumption the
expressed simply thus —

CH2— CO Br, Br.CH<

— CH2

series of

reactions would be

io-

-CO + Eton Br . CH2— CO

> — CH2 Br.CO — CH2 - HBr EtO.CO — CH2

Diketen. 7-Bromo-acetoacetyl bromide. 7-Bromo-acetoacetic ester.
If diketen were acetyl-keten, the product would be the a-brouio-
derivative, not the 7 -compound. The production of the anilide
and hydrazone-hydrazide could also be expressed in accordance
with this ring formula for diketen.

If we assume the cyclic structure for the substance, there
are three possible formulae which must be considered —
CH2— CO CH=C.OH CH=C.OH

CO — CHa CO — CH2 HO . C~ -CH

(I.) (II.) (III.)

Against the second and third of these Wilsmore adduces
the fact that diketen may be boiled with phenylisocyanate
without any interaction taking place between the two com-
pounds. This proves that the presence of hydroxyl groups in
diketen is very improbable. The refractivity of the substance
does not throw any very definite light upon the constitution.
Wilsmore l has recently determined the refractive index of the
compound, and finds the following : —

Calculated for—
(L) CHS— CO j

CO — CH. 1

18-528

18-782 0-472

(II.)CH=C.OH J
CO-CH. 1

19-542

19-728 0635

(in.) CH=C . OH j
HO.C=CH )

20-556

20-664 0-798

(IV.) CH,.CO.CH=CO

20-364

20489 0-702

Observed for diketen

20-017

20-144 0-674

1 Wilsmore, private communication to the author.

THE POLYKETIDES AND THEIR DERIVATIVES 189

The problem of the constitution of diketen must therefore
be regarded as still sub judice, but the preponderating evidence
from the chemical side points to the compound being 1, 3-
diketo- te tramethy lene.

When reduced,1 diketen yields butyraldehyde, the reaction
probably taking the course shown below —

CO — CH2 +2H HO.CH— CH2_H2o CH— CH2

| | > | | HI I

OH2— CO CH2— CO CH— CO

CH2— CO CH2— CHO

From the monoketide ethyl acetate or from diketen we
can obtain the diketide acetoacetic ester, whose properties are
so well known that it is unnecessary to recapitulate them here ;
we may therefore pass at once to the consideration of the next
polyketide derivative — dehydracetic acid.

This substance can be obtained by polymerizing keten or
diketen in benzene solution by means of pyridine or by gentle
heating; but it is best prepared by the following method.
When acetoacetic ester is heated under a reflux condenser for a
time, it loses two molecules of alcohol, and is thus converted
into dehydracetic acid, CsHsO*. This withdrawal of alcohol
may be supposed to take place in either of two ways. In the
first case, the two molecules react together to form one long
single chain, which then folds back on itself and loses a second
alcohol molecule, as shown below —

CH3.CO.CH2.CO {GET '&]• CH3CO.CH3 COOEt — *• CH-j.CO. CHj.CO.CHj-CO.CH^COOEt

-*^ CH^CO.CH,.^ CO

CH3.CO.CH,.C CO 3 || I

I 1 II I

H.C CH, H.C CH,

\6 c6

Collie's Formula.

In the second possible method of formation, one acetoacetic
ester molecule reacts in the enolic, the other in the ketonic form —

1 Wilsmore, private communication to the author.

190 RECENT ADVANCES IN ORGANIC CHEMISTRY

CH.C

CH,.C'

H.C IHJCH.CO.CH,

H.C

H.CO.CH3

Feist's Formula.

It will be seen at once that the two methods do not lead to
identical results ; the first, proposed by Collie,1 yields the lactone
of tetracetic acid ; while the second, brought forward by Feist,2
leads to the same nucleus, but produces two side chains instead of
one. The Collie formula fails to explain the presence of phenyl-
methyl-pyrazyl-phenyl-methyl-pyrazolone as a bye-product in
the interaction of phenyl-hydrazine and dehydracetic acid ; 3
while the Feist formula cannot be brought into agreement with the
fact that with phosphorus pentachloride dehydracetic acid reacts
as if it contained two hydroxyl groups.4 Both formulae represent
the properties of the compound almost equally well, and so far it
has been impossible to say which is the correct one. In the fol-
lowing pages the Collie formula will be adopted, as its decompo-
sitions are more easily represented than those of the Feist formula.

It will be noticed that the polyketide series which we have
described is not quite complete, as we have omitted to deal
with triacetic acid. This omission we can now remedy. When
dehydracetic acid is heated with ninety per cent, sulphuric acid
the ring opens, acetic acid is split off, and the ring closes again,
forming triacetic lactone 5 —

,\ ?H

CH3.CO.CH2.C

CH3.CO.CH2.C CO

COOH CH3.CO.CH2.CO COOH

H.O CH?

OH
CH3.C COOH

H.C; CH

1 Collie, Trans. Chem. Soc., 1891, 59, 179.

2 Feist, Annalen, 1890, 257, 253.

4 Collie, Trans. Chem. Soc., 1891, 69, 179.

H2C CH2
CO.OH^H
III

3 Benary, Ber., 1910, 43, 1070.
6 Ibid., 1891, 59, 617.

THE POLYKETIDES AND THEIR DERIVATIVES 191

From triacetic lactone we can reproduce dehydracetic acid by
the action of acetic anhydride in presence of sulphuric acid.

This production of triacetic lactone, however, is not the only
reaction brought about by the action of sulphuric acid. If we
dilute the sulphuric acid a little, using eighty-five per cent,
instead of ninety per cent, strength, the action takes quite a
different form.1 As in the previous case, water is added on,
the ring opens, and tetracetic acid is formed ; but instead of
breaking down into triacetic and acetic acids, the substance
enolizes in a new place and again loses water to form a new
acid —

OH HO

I I

CH3.CO CO.CH2.COOH CH3.C C.CH2.COOH

II II I!

CH2 CH2 > H.C C.H

\ / \ /

CO CO

Tetracetic acid. Enolic form.

CH3.C C.CH2.COOH

II II

H.C C.H

\ /

Isomer of dehydracetic acid.

This sensitiveness to very slight variations in the reagents used
is typical of the polyketide series, as has already been pointed
out in the case of the action of weak and stronger alkalis upon
diacetylacetone.

The action of concentrated hydrochloric acid, again, differs
from those of the two sulphuric acid solutions we have just
described. Boiling concentrated hydrochloric acid converts
dehydracetic acid into dimethyl-pyrone, the reaction probably
taking place in the following way : —

1 Collie and Hilditch, Trans. Chem. Soc., 1907, 91, 787.

192 RECENT ADVANCES IN ORGANIC CHEMISTRY

CH3.CO.CH3.C CO

CH3.CO.CH,.

H-HoO .

COOH CH3.CO.CHa.CO

-CO2 v_

OH HO
CH3.C C.CH3

C.H

H.C

CH, CH,
CO

When the dimethyl-pyrone thus obtained is analyzed, how-
ever, it is found to have the composition CvHgC^Cl, which cor-
responds to a compound of one molecule of dimethyl-pyrone
with one molecule of hydrochloric acid. The substance is not
a chlorine substituted pyrone derivative, but behaves exactly
like the hydrochloride of an organic base. Collie and Tickle,1
who were the discoverers of this class of substance, prepared
a series of compounds of dimethyl-pyrone with many of the
common acids, both organic and inorganic, as well as metallic
double salts, and from a study of their properties drew the con-
clusion that the oxygen atom which forms the bridge in the
pyrone nucleus has basic properties akin to those of a tertiary
nitrogen atom. Thus, just as tertiary amines form ammonium
salts, divalent oxygen compounds may unite with acids to form
" oxonium salts." The compound of dimethyl-pyrone with
hydrochloric acid would on this hypothesis be represented by
the formula —

H Cl

\ /

CH3.C C.CH3

II II

H.C C.H

V : .„•' ,-..,

1 Collie and Tickle, Trans. Chem. Soc., 1899, 75, 710.

THE POLYKETIDES AND THEIR DERIVATIVES 193

Dimethyl-pyrone is a white crystalline solid, subliming at
low temperatures and easily soluble in most organic liquids.
With acids it forms well- crystallized salts, soluble in, but
hydrolyzed by, water. Though it contains a carbonyl group, it
does not react with either hydroxylamine or phenylhydrazine.
This peculiar behaviour has led Collie * to put forward the view
that not one but both the oxygen atoms in the pyrone nucleus
are quadrivalent in the oxonium salts ; while in the base itself
one oxygen atom is supposed to be always quadrivalent. On
this view the formulae of dimethyl-pyrone and its hydrochloride
would be written thus —

CH3.C C.CH3 CH3.C C.CH3

H.C

| H-0-C1H
/

This view of the pyrone structure is supported to a certain
extent by an examination of the refractive indices of pyrone
derivatives which has been carried out by Miss Homfray.2 In
both of the above formulae the peculiar resemblance to the
benzenoid type is manifest, and Collie has been led to suggest
that the root-substance of the pyrone class has a structure which
resembles that of pyridine. To this hypothethical compound
he has given the name " oxene" 3 as the compound is the oxygen
analogue of benzene and pyridine.

H H

C.H H.C C.H

N O

S\ S\ S\

HC CH HC CH HC CH

I II I II I II

HC CH HC CH HC CH

*/ V

c c

Benzene. Pyridine. Oxene.

1 Collie, Trans. Chem. Soc., 1904, 85, 971; cf. Willstatter and Pummerer,
Ber., 1904, 3733 ; 1905, 38, 1461.

2 Homfray, Trans. Chem. Soc., 1905, 87, 1443.
1 Collie, Trans. Chem. 8oo.t 1904, 85, 971.

194 RECENT ADVANCES IN ORGANIC CHEMISTRY

This view lias certain advantages, for upon the ordinary
formula it is difficult to explain the comparative stability of
the pyrone compounds.

Before dealing with the hydration product of dimethyl-
pyrone, mention may be made of a substance1 which stands
midway between the pyrone and benzene series. On treatment
with dilute alkalis, dimethyl-pyrone is converted into an isomeric
body which appears to have the following constitution : —

CH2:C C.CH3

I II

H.C C.H

When this substance is boiled with acids it is converted
into the corresponding salt of dimethyl-pyrone. This change
involves only the wandering of a hydrogen atom from the
oxygen to the methylene group. On treatment with strong
alkali the substance undergoes a more complicated isomeric
change and yields orcinol. The steps involved in this reaction
are probably the following : —

0 OH HO

CH2:C C.CH3 +H20 CH2 = C C . CH3

I II -- > | II

HC CH HC CH

OH OH

CH2HO CH

s \ s \

HO— C C— CH3 _H2o HO-C C— CH3

I II -- > | II

HC CH HC CH

NX c

OH OH

1 Collie and Stewart, unpublished observation.

THE POLYKETIDES AND THEIR DERIVATIVES 195

Collie 1 has prepared the corresponding diacetyl derivative,
which behaves in a similar manner ; with acids it is converted
into the salt of diacetyl-dimethyl-pyrone, while alkalis change
it to diacetyl-orcinol.

We must now turn to the substance which is obtained by
the addition of one molecule of water to dimethyl-pyrone. The
action requires the presence of alkalis, and is best carried out
by boiling dimethyl-pyrone with a strong solution of barium
hydrate. After neutralizing the excess of alkali, the solution is
shaken out with ether, by which means diacetyl-acetone is
extracted. The course of the reaction involves the formation
and decomposition of the barium salt of diacetyl-acetone —

0 0— Ba— 0

CH3.C C.CH3 CHg.C C.CHg

II II > II II

H.C C.H H.C C.H

v > v

OH HO ^

CHg.C C.CHg CHg.CO CO.CHg

II II — > I I

H.C C.H CH2 CH2

\ / \ /

CO CO

Diacetyl-acetone forms colourless mica-like crystals, which
volatilize at ordinary temperatures. It is unstable, losing
water with great ease, and changing into dimethyl-pyrone;
while, under certain conditions, it breaks down into acetone
and acetic acid. It forms one of the very small class of tri-
ketones, and with it we reach the highest stable member of the
polyketide class.

At the beginning of this chapter we called attention to the
view that many of the simple substances found in plants were
not the results of direct synthesis, but rather of synthesis
followed by decomposition, and in diacetylacetone we have a
substance which will serve as a typical example of this method.
In the first place, by spontaneous dehydration at ordinary

1 Collie, Trans. Chem. 800., 1904, 85, 971.

196 RECENT ADVANCES IN ORGANIC CHEMISTRY

temperatures, we get dimethyl-pyrone. By using acid dehy-
drating agents we can form orcinol l —

CH3 CH

/ / \

CO 0 = 0— CH3 CO C— CH3

I I > \ I

CH2 CH2 CH2 CH2

XCOX CO

-CH3

Stronger dehydrating agents produce scarlet dye-stuffs,2 which
are probably similar to those obtained by dehydrating di-
methyl-pyrone. We have already described the formation of
the pyridine, benzene, naphthalene, and isoquinoline deriva-
tives —

CH3.C C.CH3 CH3/XCH2COCH3

II II

H.O O.H HJCO.CH2COCH3

CO OH

Pyridine compound. Benzene compound.

H H
C C

CH3.C C C.CH3

II I I
H . C C C . CO . CH3

\/-w

C C

OH OH OH CH2.COCH3

Naphthalene compound. Isoquinoline compound.

1 Collie and Myers, Trans. Chem. 8oc., 1893, 63, 122.

2 Collie and Stewart, unpublished observation.

THE POLYKETIDES AND THEIR DERIVATIVES 197

There is one polyketide derivative which we have not yet
mentioned. If we treat keten with hydrochloric acid and with
ethyl alcohol we get acetyl chloride and ethyl acetate ; from the
latter we can produce acetoacetic ester, and thence by the aid
of the acetyl chloride we can synthesize acetylacetone. This
substance completes the series of ketones which we have derived
from the simple keten group, and it may be well to give a table
showing the relations of each member to the others.

Ketens.

Acids.

Ketonee.

Keten

(CH2 : CO)

Diketen
(CH2 : C0)2

Acetic
H.(CH2.CO).OH

Acetoacetic
H . (CH2 . CO)2OH

Acetone
H.(CH2.CO).CH3

Acetylacetone
H . (CH2 . C0)2 . CH3

Triacetic
H.(CH2.CO)3OH

Tetracetic
H.(CH2.COXOH

Diacetylacetone
H.(CH2.CO)3.CH3

We may subjoin another tabular statement (see next page),
which brings out the relations between dehydracetic acid, the
pyrones, and the aromatic series.

So far, we have described only those polyketide derivatives
which can be obtained from keten by methods which have
actually been worked out experimentally. There is one most
important class of substances, however, which do not come
within this category, though, theoretically, they belong to the
polyketide derivatives. The sugars — though we at present
have no means of synthesizing them from keten, or any of its
simple derivatives — are very closely related to the keten group.
Collie 1 has indicated the lines which should be followed in such
syntheses ; but at present the proper conditions have not been
discovered.

Wills tatter and Pummerer 2 have shown that when we act
upon pyrone with metallic alcoholates bishydroxymethylene-
acetone derivatives are produced. If we consider the effect of
opening the pyrone ring with a water-molecule instead of a

1 Collie, Trans. Chem. 8oo.t 1907, 91, 1806.

2 Willstatter and Pummerer, Ber., 1905, 38, 1461.

198 RECENT ADVANCES IN ORGANIC CHEMISTRY

£ s

I-

g" §

g c 2 o g

"0 " ow d s o >*
c c-aJa'3 « S g.

DO 000

tug-* °-0-$0*£>5i£3
-o u 3^ w_a_ a a^ 2 222

"ocT OO UU(JU '^'^'^

glff

t
tm I

• '

00 ga;""

1 i T + tt +«5s+:

oo 00% or°26'

3 BO"

-o 5 °-

ucj5 ooSi^ u"ou"

THE POLYKETIDES AND THEIR DERIVATIVES 199

molecule of alcoholate, we find that bishydroxyinethylene-
acetone itself (2) would result. If this could be converted
into the isomeric form (3) by the wandering of a hydrogen
atom, and the resulting compound could be induced to combine
with two molecules of water at the double bonds, the pentose
(4) would be produced. So far, no successful attempt has
been made to produce this change, but if the proper conditions
could be found there seems no reason why it should not be
carried out.

0
H.C C.H HO. C.H H.C.OH

H.C C.H H.C C.H

\co/X

V x

(1.)

HO . C . H ti^ HO . CH2

II I

C.H HO. C.H

C.OH H.C.OH

C.H HO . C . H

I I

CO CO

i . 1

(3.) (4.)

By successive hydration and dehydration a different type of
product would result —

, io

iSb

CH2

C.H

HO. C.H

• i°

HO.C

HO. C.OH

<!o

C.OH

1
H . C . OH

H.C

H.C.H

CH2

(I-)

(II.)

(III.)

(IV.)

(V.)

(VI.)

200 RECENT ADVANCES IN ORGANIC CHEMISTRY

If we take as our starting-point the group (I.) and convert
it into the enolic form (II.), we can then add a molecule of
water on to the double bond to form (III.) This substance
could then be dehydrated to produce (IV.), to which water
might be again attached, giving (V.), in which two hydroxyl
groups are attached to the same carbon atom. This compound
would lose a molecule of water, leaving (VI.).

A comparison of the formulae (I.) and (VI.) shows that the
whole process implies a wandering of the hydrogen atoms to
the lower end of the chain, and a corresponding migration of
the oxygen atoms to the other. This purely theoretical series
of actions could then be repeated, and the final result would be
a loss of carbon dioxide from one end of the chain, and a
building up of an aliphatic chain at the other end. Some
such process may take place in the living organism during the
formation of oils or fats, and the liberation of carbon dioxide in
respiration would be explicable in the same way.

We have now completed our survey of the polyketides and
their derivatives, and in conclusion we may point out the
salient features of the classes with which we have dealt. The
polyketens themselves, (CH2 : C0)n, are remarkable chiefly for
their great reactivity ; they are easily attacked by any ordinary
reagents, and further possess the power of polymerization to a
marked degree. Their union with acid or neutral substances
produces compounds which in turn are reactive, though not to
the same extent as the parent bodies ; but if, on the other
hand, the polyketens be combined with basic substances, the
products are not at all reactive. The higher members of the
polyketide group when combined with water tend spontaneously
to lose carbon dioxide, and become converted into ketonic com-
pounds of a lower series, which in turn may be dehydrated
to form benzenoid or pyrone derivatives, both of which are
comparatively stable. Thus these substances as a class illus-
trate the dual tendencies at work in the whole field of organic
chemistry — the synthetic and the analytic ; the simpler, more
reactive group tending always to attract other atoms and form
more complex derivatives, while these in turn become unstable
and break down into new and more stable forms.

CHAPTEK IX

THE QUINOLES

1. Introductory.

IF in the formula of benzoquinone (I.) we replace one of the
carbonyl radicles by a secondary (II.) or a tertiary (III.)
alcohol radicle, we shall obtain the structural formulae of two
new types of compounds which have been termed quinoles by
Bamberger.1 Compounds of the type (II.) are secondary
quinoles ; those of the structure (III.) are tertiary compounds.

0 H OH R OH

II \ / \ /

c c c

H.C C.H H.C C.H H.C C.H

II II II II II II

H.C C.H H.C C.H H.C C.H

\y \/

c c

II II II

000

(I.) (II.) (III.)

An examination of the second formula will show that it is the
mono-ketonic form of ordinary hydroquinone.

Derivatives of these substances had long been known in
the anthracene series, oxanthranol being the simplest of them :

H OH
CH C CH
H/VVX

i ii ii i

HC C C CH

/\/\^
H C CH

o - -

Oxanthranol
1 Bamberger, Ber., 1900, 33, 3607.

202 RECENT ADVANCES IN ORGANIC CHEMISTRY

but it was not until 1895 that mono-nuclear quinoles were
discovered by Zincke,1 and with this research the history of
the quinoles proper begins; for in the anthracene derivatives
the specific character of the quinole nucleus is to a great
extent concealed owing to the complications introduced into
the molecule by the presence of the two benzene nuclei.

For many years Zincke had been engaged upon the study of
the action of chlorine upon various aromatic substances; and
his results showed that this reagent was capable of converting
the stable and comparatively inert benzenoid nucleus into a
most varied and reactive series of products. One example will
suffice : the action of chlorine upon catechol.2 The action
takes place in the following stages. The first product is
tetrachloro-o-quinone (II.), which is then further acted upon
by chlorine to give hexachloro-0-diketo-K-hexen (III.). When
this substance is heated with water, it undergoes intramolecular
change, the six-membered ring being converted into a five-
membered one, and hexachloro-K-penten-hydroxy-carboxylic
acid (IV.) is produced. Oxidation with chromic acid breaks off
the side-chains and the ketone hexachloro-E-penten (V.) results,
which on treatment with caustic soda opens up, yielding
perchloro- vinyl-aery lie acid (VI.), from which ethylidene-
propionic acid (VII.) may be obtained by reduction.

Cl 01

A A

OH 01. C CO Cl.C CO

Cl.C CO C]2C CO

0 C

I /\

Cl Cl Cl

(I.) (II.) (III.)

1 Zincke, Ber., 1895, 28, 3121.

2 Zincke, Ber., 1888,21, 2719; 1889,22,486; 1891,24,908; 1893,26,2104
1894, 27, 3364.

THE QUINOLES 203

COOH

GIG

Cl Cl
(IV.)

Cl
C
Cl . C COOH

Cl.C

C.C12
(VI.) (VII.)

When Zincke applied this reaction to para-cresol1 he found
that the substance was unexpectedly stable, for the end-
product of the chlorination process still contained two nuclear
hydrogen atoms unreplaced by chlorine, CH3 . CeH2 . C130. Ee-
placing para-cresol by para-toluidine, he obtained as the end-
product of the reaction the substance CH3 . CeH2 . C170. Zincke
formulates the reaction in the following manner : —

CH3 CH3 CH3

I I I

H.C C.H

I II
H.C C.H

(I.)

# \

H.C C.H

H.C C.

1 II
Cl . C C . Cl

C12C C.
\ /

\ /

NH2

(II.)

(III.)

Bamberger, Ber., 1900, 33, 3607

204 RECENT ADVANCES IN ORGANIC CHEMISTRY

Cl CH3 Cl CH3

\X \/

c c

C1H.C C.HC1 C1H.C C.HC1

II II

Cla.C C.Cla Cla.C C.C12

v v

ii ii

NH 0

(IV.) (V.)

The difference between the reactions of the phenol and amido-
compound is to be ascribed to the fact that the intermediate
imide formed from the toludine is more reactive than the
corresponding keto-chloride which is produced from the cresol.
Now if we reduce the keto-chloride (V.) we obtain tetra-
chloro-para-cresol :

Cl . C C . Cl

I II

Cl.C C.C1

When this was treated with nitric acid, Zincke expected, by
analogy with the results obtained by him in other cases, that an
ortho-quinone of the formula CH3 . C6C1302 would be produced ;
but instead of this substance he obtained a compound having
the formula CH8 . C6HC1302. On reduction, this new substance
regenerated tetrachloro-para-cresol, and on the grounds which
we shall deal with in a later section, Zincke ascribed to it the
formula —

THE QUINOLES 205

HO CH3

Cl . C C . 01

II II

01 . 0 0 . 01

II
o

It belongs to the quinole series, and in this way a new line
of research was opened up.

2. Methods of Preparing Quinoles.

The quinoles which have been prepared up to the present
may be divided into three classes : those with no substituents
attached to the nucleus ; quinoles whose nuclear hydrogen atoms
have been replaced by alkyl radicles ; and, finally, halogen-
substituted quinoles. Since in some instances it is impossible
to prepare the parent substances by reactions which answer
quite well in the case of derivatives, it will be most convenient
to classify the methods of preparation of these substances
according to the end-products of the reaction ; and we shall
therefore take up in turn the three types of quinoles.

I. Simple Quinoles. — These were obtained by Bamberger x
in the following manner. If we subject a para-alkylated aryl-
hydroxylamine derivative to the action of dilute sulphuric acid
or alum solution, the first product at ordinary temperatures is
an imido-quinole ; but should the reaction be prolonged, this
substance breaks down into ammonia and the corresponding
quinole. If the action be not brought to an end at this point,
intramolecular change occurs and an alkyl-substituted hydro-
quinone is formed. For instance, in the case of para-tolyl-
hydroxylamine the reaction takes place in the stages illustrated
by the formulae below.

1 Bamberger, JBer., 1900, 33, 3615.

206 RECENT ADVANCES IN ORGANIC CHEMISTRY

CH3 HO CH3 HO CH3 OH

v v i

HC CH HC CH HC CH HC C CH,

I II -> II II -» II I . •* I I
HC CH HC CH HC CH HC CH

V V V V

I II II I

NH.OH NH O OH

II. Alkyl-sulstituted Quinoles. — There are three methods of
preparing this set of derivatives. In the first place, we may
apply the aryl-hydroxylamine reaction described above, starting
in this case with a di-substituted aryl-hydroxylamine instead
of a mono-substituted one. For instance, in the case of 2, 4-
dimethyl-phenylhydroxylamine l we have the following series
of changes, resulting in the production of dimethyl-quinole :—

HO CH3

XCH

C.CH3 HC C.CH3

NH.OH NH 0

Caro's reagent applied to di-substituted phenols2 produces a
certain amount of the corresponding quinole; but the yields
are small. Of course in both the aryl-hydroxylamine and the
phenol which are used as starting-points for the foregoing re-
actions one of the substituent alkyl radicles must be in the
para-position to the hydroxylamine residue or to the hydroxyl
group.

A third method3 of preparing alkyl-substituted quinoles
consists in the application of Grignard's reagent to substituted

1 Bamberger, Ber., 1900, 33, 3647.

2 Ibid., 1903, 36, 2028.

3 Bamberger and Blangey, Ber., 1903, 36, 1626.

THE QUINOLES 207

quinones. Benzoquinone itself does not give a quinole in this
way, but when toluquinone is treated with magnesium methyl
iodide, a dimethyl-quinole is produced : —

0 HO CH3

II \'/

c c

HO CH HO CH

HO C.H3 HC C.CH3

v -• v

II II

O 0

III. Halogen-substituted Quinoles. — In the introductory
section of this chapter we have seen that these bodies may be
prepared by the chlorination of para-substituted aromatic
amines. The researches of Zincke1 and of Auwers2 have
shown that when halogen-substituted phenols are heated with
concentrated nitric acid, the corresponding quinoles are pro-
duced. In this case the intermediate products are cyclic nitro-
ketones, so that the reaction takes the following course : —

CH3
C
Br.C C.Br

N02 CH3

\/
C

Br.C C.Br

HO CH3

»•; )<

Br.C C.Br

--> II II
Br.C C.Br

Br.C C.Br

II
0

Similar results are obtained when halogen-substituted phenols
are dissolved in glacial acetic acid and then treated with nitrous
fumes.

When bromine is allowed to act upon para-cresol, Zincke
and Wiederhold 3 found that there are three possible types of

1 Zincke, Ber., 1895, 28, 3121 ; 1901, 34, 253, J. pr. Ch., [2] 1897, 56, 157.

2 Auwers, Per., 1897, 30, 755; Annalen, 1898, 302, 153.

3 Zincke and Wiederhold, Annalen, 1901, 320, 199.

208 RECENT ADVANCES IN ORGANIC CHEMISTRY

reaction product, one or another preponderating according to
the conditions of the experiment. If the bromine acts upon
the phenol in chloroform solution at ordinary temperatures,
we obtain simply the mono-, di-, and tribromo-derivatives of
paracresol : —

CH3

^
C.Br

Br.c"

C.Br

II
r-C

C.Br2

II
Br.C

C . Br,

When water is substituted for chloroform, we get normal keto-
bromides, such as : —

CH3

/ \
HC CH

II I
Br.C C.Br2

CO CO CO

But if we allow bromine in excess to act directly upon the
phenol, without any solvent, in a sealed tube at 100° C. for
several hours, bromine enters the side-chain, and we get tetra-
bromo-^-cresol-pseudobromide : —

H CH2Br

Br . C C . Br

II II
Br . C C . Br

; V

When this substance is boiled with nitric acid (sp. gr. 1*4) it
yields pentabromo-toluquinole, which has the following con-
stitution :

THE QUINOLES 209

HO CH2Br

Br.C

.Br

II
Br.C

II
C

.Br

3. The Properties of the Quinoles.

As a class, the quinoles are distinguished by extreme re-
activity and lability of structure. They react with many of
the ordinary reagents with great ease ; and in addition they
are capable of undergoing very far-reaching isomeric changes.
With the latter division of their properties we shall deal in a
special section; at present we shall confine ourselves to the
action of the commoner chemical reagents upon compounds of
the quinole type.

In their general chemical character, the quinoles resemble
weak acids. They are soluble in caustic alkali, but not to any
great extent in sodium carbonate solution. From this beha-
viour we may deduce the presence of a hydroxyl group in their
structure.

When reducing agents such as sulphurous acid, zinc dust
and acetic acid, or zinc and aqueous ammonium chloride are
employed upon quinoles, reduction takes place with great ease.
Bamberger 1 has found that even ferrous sulphate and sodium
carbonate solution are sufficient to reduce a dimethyl-quinole
to the corresponding phenol. This formation of the benzenoid
nucleus in preference to the quinonoid one is, as we shall see
later, one of the most characteristic features of the intramole-
cular changes which have been observed in the quinole group.
The tendency is so strong that reduction is brought about by
such unlikely substances as hydrobromic acid and phosphorus
pentabromide.2 In all these cases we may formulate the reduc-
tion process as a true reduction in one of its stages, the next
stage being the loss of water or hydrobromic acid. In the

1 Bamberger, Ber., 1900, 33, 3616.
8 Auwers, Ber., 1902, 35, 445.

210 RECENT ADVANCES IN ORGANIC CHEMISTRY

reduction by means of pentabromide of phosphorus, the first
step appears to be the replacement of the hydroxyl group of
the quinole by a bromine atom ; the hydrobromic acid thus
liberated then reacts as usual with the quinole. The following
formulae give some idea of the reduction of methyl quinole by
zinc dust, hydrobromic acid, and phosphorus pentabromide.

Zinc dust and acetic acid—

HO CH3 HO CH3

CH3

X X

HC CH H2 HC CH -H2o

HC CH

II II > II II >

1 II

HC CH HC CH

HC CH

V/ \/

\ /

C C

0 HO H

1
OH

Hydrobromic acid—

HO CH3 Br CH3 CH3

\y \/ i

C C C

/\ /\ ^\

HC CH HC CH HC CH

II II > II II > 1 II •

•j- Br2 + H20

HC CH HC CH HC CH

\ / \ / <^. /

V V V

II / \ 1

/ \

0 HO Br OH

Phosphorus pentabromide —

HO CH3 Br CH3 Br CH3

CH3

\/ \/ \/

C C C

HC CH HC CH HC CH

HC CH

II II -> II II -> II II ->

1 II

HC CH HC CH HC CH

HC CH

V V V :

\ /

II II / \

0 0 H OH

THE QUINOLES 211

From their behaviour towards alkalis, it was to be expected
that the quinoles would yield acyl derivatives ; and this has
proved to be actually the case. Mono-acetates are produced
when quinoles are treated with acetic anhydride or acetyl
chloride ; and the benzoyl derivatives are formed in the usual
way by the Baumann-Schotten reaction. It has been found
that in the case of the quinoles derived from brominated
phenols, the acyl derivatives are even more easily reduced
than the parent quinoles are. Auwers l mentions as an example
of this the case of tribromo-m-xyloquinole. This substance is
not attacked at ordinary temperatures by hydrobromic acid in
acetic acid solution ; reduction begins only at 100° C. But
when the acetyl derivative of the quinole is subjected to the
same agent, it is reduced even at ordinary temperatures to
tribromo-xylenol.

CH3CO.O CH3 CH3

\/ I

C C

/\ ^\

Br.C C.Br Br.C C.Br

II II > I II

Br.C C.CH3 Br.C C.CH3

It is easy to see that this power of reduction which hydro-
bromic acid possesses must exert considerable influence in the
case of the action of acetyl bromide upon quinoles. While
acetyl chloride yields the ordinary mono-acetate of the quinole
employed, it is frequently found that acetyl bromide produces
the acetate of the corresponding phenol ; so that in the latter
case the simple acetylation has been complicated by the
reduction of the quinole to the phenol by means of the
hydrobromic acid liberated in the course of the reaction.
The reducing action of the hydrochloric acid which is liberated
when acetyl chloride is used is very much feebler than that of
the hydrobromic acid ; and consequently the quinole acetate in
that case is not transformed into the phenolic derivative,

1 Auwers, Ber., 1902, 35, 446.

212 RECENT ADVANCES IN ORGANIC CHEMISTRY

Turning now to the reagents which react with carbonyl
groups, we find that a difference is to be noted between two
classes of the quinoles. Those quinoles which have substituents
in the two positions ortho to the carbonyl radicle follow the
usual rule in such cases and do not interact with phenylhydra-
zine, hydroxylamine, or semicarbazide.1 The case of an unsub-
stituted quinole differs from this. Bamberger and Eudolf2
have found that when xyloquinole is treated with hydroxyl-
amine, it reacts with two molecules of the latter, giving a
substance which is both an oxime and a substituted hydroxyl-
amine.

HO CH3

CH . NH . OH

. C CH2

II
N.OH

With nitro-phenylhydrazine and with semicarbazide, the
same quinole gives cyclic substances3 having the following
formulae : —

N02.C6H4.N

-C
\

NH2CO . N-

HC C.CH3
\/

HC C . CH3

From the foregoing, the salient features of the quinoles can
be inferred, and in the following section we shall show how
the constitution of these substances may be deduced.

1 Auwers, Ber., 1902, 35, 444.

2 Bamberger and Budolf, Ber., 1907, 40, 2236.
• Bamberger, Ber., 1900, 33, 3620.

THE QUINOLES 213

4. The Constitution of the Quinoles.

In the previous section we have dealt with most of the
material which will be required to establish the structure of
the quinoles ; and in the present section it will be convenient
to deal with the question point by point.

I. The quinoles contain a ring of six carbon atoms. This
is shown by their conversion into benzene derivatives on
reduction.

II. The quinoles contain a hydroxyl group. This is proved
by their yielding acetyl derivatives ; and also by the fact that
they dissolve in alkalis more readily than in water, and that
acids precipitate them from these alkaline solutions.

III. They contain a carlonyl group. The action of phenyl-
hydrazine, hydroxylamine, and semicarbazide upon quinole
derivatives establishes this.

IV. Quinoles have a methyl radicle in the 1, ^-position to the
carlonyl group. The most convincing proof of this is an indirect
one. It was found by Brady 1 that quinoles are produced only
from phenyl-hydroxylamine derivatives which have an alkyl
group in the para-position to the hydroxylamine residue. For
example, the five compounds below gave no quinoles when
treated in the usual manner with sulphuric acid or alum
solution : —

GIL

NH.OH NH.OH NH.OH
CH3

'CH3
NH.OH NH.OH

1 See Bamberger, Per., 1900, 33, 3616.

214 RECENT ADVANCES IN ORGANIC CHEMISTRY

but all the following four compounds yielded quinoles : —
CH3 CH3 CH3 CH3

NH.OH NH.OH NH.OH NH.OH

It is clear that quinole formation takes place only when there
is a methyl group para to the hydroxylamine residue ; and
since the group — NH.OH in the aryl-hydroxylamines cor-
responds to the carbonyl radicle in the quinoles, we may con-
clude that in the latter compounds the carbonyl and methyl
radicles are in the para-position to one another.

Having advanced to this point, it is clear that there are
only three possible formulae for methyl-quinole : —
HO CH3 H CH3 H CH3

X / v

HC CH HC C.OH HC CH

HC CH HC CH HC C.OH

o o o

(I.) (II.) (III.)

The third of these is the enolic form of an ortho-diketone :

H CH3

HC CH2

II I

HC CO

and since the properties of methyl-quinole do not correspond
to those which we should expect to find in ortho-diketones,
we may dismiss this formula. If we examine the second
formula, we find that it is a tautomeric form of cresorcinol ; so

THE QUINOLES 215

we may reject this formula also. This leaves us with the first
formula as the only probable one.

Further evidence in favour of this structure is furnished
by the formation of quinoles from quinones by means of the
Grignard reaction : —

CH3 OMgl CH3 OH

X " v

CH3.C C.H cH3Mgi CH3.C C.H CH3.C CH

II II > II II > || ||

H.C C.H H.C C.H H.C CH

v v y

The yields are so small that this reaction would not in itself
furnish a perfectly satisfactory proof of the quinole structure ;
but it lends additional weight to the other proofs.

5. Intramolecular Change in the Quinole Series.

In the foregoing sections we have encountered one or two
instances in which the quinole derivatives were converted into
benzenoid compounds by the action of suitable reagents; in
the present section we shall discuss several other cases of the
same type.

It has been shown by Stewart and Baly1 that the intro-
duction of substituents into the quinone nucleus tends to give
the intramolecular vibrations of the system a more and more
pronounced benzenoid character; the compounds lose their
ketonic properties to a marked extent,2 and their absorption
spectra approximate more and more closely to those of the ben-
zene derivatives. In the case of the quinoles, the replacement
of the one carbonyl group of the quinone by a tertiary alcoholic
radicle appears to produce an analogous effect ; but since in this
case there is a possibility of the substances undergoing intra-
molecular rearrangement, the process is carried a step further
than in the simple quinones, and a wandering of radicles ensues
Which actually produces true benzene derivatives.

In the quinole group, the chief part of the molecule which

1 Stewart and Baly, Trans. Chem. Soc., 1906, 89, 618.

2 Kehrmann, Ber., 1888, 21, 3315; /. pr. Ch. 1889, 39, 399; 40, 257.

216 RECENT ADVANCES IN ORGANIC CHEMISTRY

is involved in intramolecular change is the tertiary alcoholic
radicle, and in most cases the methyl group in this wanders to
some other part of the six-member ring. We may now
describe several instances in which this methyl radicle is
removed from its original position and re-attached to the carbon
atom in the ortho-position.

Bamberger1 has found that when quinoles such as those
shown below are subjected to the action of hydrogen, or, better,
hydroxyl ions, they are converted into homologues of hydro-
quinone, as the f ormulae indicate : —

HO CH3

HO CH

II II *

HO C.CH3

In these cases, it is clear that the forces which bring about
the intramolecular change are of a strength sufficient to elimi-
nate a hydrogen atom from the nucleus and to replace it by a
methyl radicle. A much more powerful action is shown in
another case, which was investigated by Zincke.2 If we treat
tetrabromo-ethyl-quinole with concentrated sulphuric acid in
the cold, we find that hydrobromic acid is eliminated, and the
ethyl radicle replaces the bromine atom which has been driven
out of the nucleus : —

1 Bamberger, Ber., 1900, 33, 3618. 2 Zincke, Sir., 1901, 34, 253.

THE QUINOLES 217

HO C2H5 0

X ;;' A- •"'•'•

Br.G C.Br -HBr Br.C C . C2H5

II II > II II

Br.C C.Br Br.C C.Br

v v •

II II

0 O

A still more extraordinary wandering is seen in a case
mentioned by Auwers.1 If we heat quinoles with acetic
anhydride, they yield monoacetates ; but if sodium acetate be
present, it is found that a diacetate is formed in the case of
those quinoles which have a methyl group in the position ortho
to the tertiary alcohol residue. Investigation shows that this
second acetyl group has entered the methyl radicle ; so that the
reaction really takes place in the following stages : (1) acety-
lation of the hydroxyl group of the quinole ; (2) wandering of
the acetate group into the methyl radicle in the ortho-position
to it ; (3) acetylation of the new hydroxyl group formed by the
change from the quinolic to the benzenoid structure : —

CH3CO.O.CH

>H CH3CO.O

Auwers, Per., 1902, 35, 449.

218 RECENT ADVANCES IN ORGANIC CHEMISTRY

If we examine the behaviour of quinoles whose methyl
group contains a halogen substituent, a fresh set of changes
is presented to us. In the first place, if we allow equimo-
lecular quantities of caustic soda and quinole to interact, we
get, as Zincke has shown,1 an oxide of the following type
formed : —

HO CH2Br

0 CH2

HC CH

II II
HC CH

II
O

4- NaOH = NaBr + H20

HC CH

II
0

On the other hand, if excess of alkali be used, and if the
quinole contains a methyl radicle in the ortho-position to the
hydroxyl, the reaction takes another turn ; for the oxide reacts
with the solvent (methyl or ethyl alcohol) with the following
results : —

0 CH2 CH30 CH2OH

\/ \/

C C

HC C . CH3 + CH3OH HC C . CH3

II II > II II ->

HC CH HC CH

\/ \/

C C

CH2OH

CH20 , CH

In special cases it has been observed that the methyl
radicle remains attached to its carbon atom, while another
group wanders. For example, Auwers 2 finds that in the

1 Zincke, Ber., 1895, 28, 3121 ; Annalen, 1901, 320, 177 ; Auwers, Ber., 1902,
35, 451.

2 Auwers, Ber., 1902, 35, 454.

THE QUINOLES 219

nitro-ketones, the nitro-group wanders sometimes into the
ortho- and sometimes into the meta-position, as shown in the
following formulae : —

CH3

\/
C

HC C . CH3

II II
CH3.C C.N02

II
0

N02 CH3

\/
C CH3

II II >

BrC C.Br

II
O

A similar wandering of the quinole hydroxyl group to the
meta-position has been observed in one case.1 When xylo-
quinole is treated with alcoholic sulphuric acid, it is converted
into a mixture of substances indicated in the formulae below,
from which it will be seen that in the one case the methyl
radicle has wandered, while in the second case the hydroxyl
group has changed its position in the ring.

1 Bamberger, Per., 1907, 40, 1895.|

220 RECENT ADVANCES IN ORGANIC CHEMISTRY

OH.—

In conclusion, we may add two instances in which the
groups in the para-position to one another are acted upon
simultaneously. In the first place we may take the action of
boiling acetic anhydride upon the bromo-substituted quinoles in
which the bromine atom is in the side-chain. In this case * a
loss of hydrobromic acid takes place, but at the same time
reduction occurs, and the final product is the diacetyl deri-
vative of para-hydroxy-benzyl alcohol : —

HO CH2Br

HO CH

CH3CO . 0 . CH2

HC CH

(CH3CO)20 =

CH3CO

+ HBr + 0

Again, Bamberger 2 has found that in some cases the action of
phenylhydrazine ends in the production of azo-compounds, the
intermediate step being the elimination of water between the

Auwers, Ber., 1902, 35, 450.

Bamberger, Ber., 1902, 35, 1426.

THE QUINOLES

221

hydroxyl group of the quinole and the hydrogen of the imino-
radicle of the hydrazine nucleus : —

HO CH3

HC CH

II
0

6. Conclusion.

From the changes dealt with above, the reader will have
gathered something of the extraordinary lability of the quinole
nucleus. There is hardly another class of compounds which
shows such examples of intramolecular interchange of groups ;
and at present the only determining factor which we can detect
underlying the quinole rearrangements appears to be the
attempts of the unstable quinoles to revert to the more stable
benzenoid type. It seems clear that further investigation of
this field might lead us to an understanding of the cause of the
stability which the aromatic series in general exhibits towards
most reagents. In the quinoles we apparently have a bridge
between the alicyclic compounds on the one hand and the
aromatic bodies on the other; and as is the case with most
intermediate compounds, the quinoles are more reactive than
members of either the aliphatic or the aromatic series,

CHAPTER X

THE TRIPHENYLMETHYL QUESTION

1. Introductory.

ANY one who glances through the journals of the chemical
world for the last few years must be struck by the enormous
production of new compounds which is at present going on;
and if he reflects at all, he will be driven to ask himself what
criterion should be applied in order to distinguish the really
important substances from what we may term the by-products
of synthetic chemistry. It is perfectly clear from our experience
that the only fate which can overtake the majority of these new
compounds is that their dossiers will be "neatly tucked away
in Beilstein, the Abstracts published by the various Chemical
Societies, or in other equally convenient depositories of infor-
mation." There they will remain at best in a dormant condition,
waiting the time when some Analogie-arbeit necessitates a
knowledge of their properties. On the other hand, those new
bodies which have any interest apart from their melting-points
soon become centres of new research ; and the more important
of them usually lead to investigations extending far beyond the
constitution and properties of the original compound. For
example, the researches which more than a generation ago took
their rise in the constitution of acetoacetic ester have not yet
reached their final stages.

This ramification of interest has seldom been so strongly
marked within recent years as in the case of the substance
termed triphenylmethyl ; and it is the rapid extension of the
field of research in this division of the subject which makes
any treatment of the triphenylmethyl problem difficult. In the
present chapter, it will be necessary to confine ourselves as far as
possible to the narrow question of the constitution of triphenyl-

THE TRIPHENYLMETHYL QUESTION 223

methyl and only to touch lightly upon the wider questions
which are closely bound up with it.

The discovery of triphenylin ethyl took its rise in an attempt
to prepare hexaphenyl-ethane, which was made by Gomberg 1
in 1900. He allowed " molecular " silver to act upon triphenyl-
bromo-methane, and obtained a compound which he naturally
supposed to be hexaphenyl-ethane, for the reaction would
normally have taken the course expressed in the formulae
below —

2 C6H5— C— Br + 2Ag = 2AgBr + C6H5— C— C— C6H5

Celts Cells

On analysis, however, the substance was found to have about
six per cent, too little carbon to agree with the hexaphenyl-
ethane formula ; and further examination showed that it could
not be a hydrocarbon at all, but must contain oxygen.

This oxygen might have been introduced in either of two
ways : it might have been imported through the silver used in
the reaction ; or it might have been derived from the air. The
experiments were therefore repeated, other metals, such as zinc
and mercury, being used instead of silver ; and still the resulting
substance was found to be oxygenated. From this it was clear
that atmospheric oxygen was the origin of the oxygen in the
end-product; and further experiments were made in which
precautions were taken to exclude air from the apparatus. The
end-product in this case differed from that which had previously
been obtained; and on analysis it was found to have the
composition corresponding to hexaphenyl-ethane.

An examination of its properties, however, brought Gomberg
to the conclusion that the substance which he had obtained
could not be hexaphenyl-ethane ; for he had expected that that
body would be an extremely stable compound, whereas his
synthetic hydrocarbon was very reactive.

At this point we may give a resume of the chief properties
of the hydrocarbon. When first prepared, it is a colourless
crystalline solid, which dissolves with great readiness in most

1 Gomberg, /. Amer. Chem. 8oc.t 1900, 22, 757; Ber., 1900, 33, 3150.

224 RECENT ADVANCES IN ORGANIC CHEMISTRY

organic solvents, giving yellow solutions. Even at zero it
reacts with iodine to form triphenylmethyl iodide. Exposure
to the air even for a short time is sufficient to transform it into
a peroxide ; and Gomberg 1 has been able to prove that this
same peroxide can be produced by the action of sodium peroxide
on triphenyl-chloro-methane (but not by the spontaneous
oxidation of triphenylmethyl chloride or of triphenyl car.binol
under the same conditions). From this we may deduce that
the peroxide has the constitution —

C6H5 C6H5

C6H5— C— 0— 0— C— C6H5

The hydrocarbon forms double compounds2 with ethers,
esters, ketones, nitriles, or aromatic hydrocarbons (and amylene),
the composition of these substances corresponding to one mole-
cule of ether (or of the other substances) plus one molecule
of hexaphenyl-ethane. Gomberg ascribed the formation of the
oxygenated derivatives to the change of the oxygen from the
divalent to the quadrivalent condition, and formulated the con-
stitution of the substances generally as derivatives of the
following types : —

K C(C6H5)3 E C(C6H5)3 K C(C6H5)3

V \o( ):0/

R C(C6H5)3 K C(C6H5)3 EO C(C6H5)3

C(C6H6)3

E.C=NX

C(C6H5)3

The fact that these substances are actually compounds and
not simply mixtures in which the ether or other body is
held mechanically is proved by the fact that similar compounds
are formed with carbon disulphide and chloroform, and these

1 Gomberg, Ber., 1900, 33, 3150.

2 Ibid., 1905, 38, 1333, 2447.

THE TRIPHENYLMETHYL QUESTION 225

latter bodies can be heated to 110°C, in a stream of carbon
dioxide without giving up their full content of chloroform or
disulphide.

There is one further point to which we must draw attention,
though it does not directly concern the hydrocarbon. It has
been shown l that the halogen salts, such as triphenylmethyl
chloride, (C6H5)3C . 01, and triphenylmethyl bromide, (C6H5)3C.Br,
when dissolved in solvents such as liquid sulphur dioxide
which have strong dissociating power, have conductivities very
nearly equal to that of methylamine hydrochloride. This
proves that in the yellow solutions obtained in this way, the
compounds are split up into two ions, one of which must be
(C6H5)3C.

From the data which we have given in the preceding
paragraphs, it is clear that the problem of the constitution of
Gomberg's synthetic hydrocarbon opens up a wide field for
speculation ; and numerous attempts have been made in recent
years to discover the solution. Four views have at one time
or another gained a certain amount of support, and we shall
deal with these in turn in the succeeding sections of this
chapter.

2. The Trivalent Carbon Hypothesis.

The reactions of his synthetic hydrocarbon — which we may
for the sake of convenience term triphenylmethyl — led Gom-
berg 2 to put forward the view that the substance contained one
carbon atom attached to three phenyl radicles, but having no
fourth radicle attached to it : —

— C

The fourth valency of the carbon atom may be supposed to be
free, or to be absorbed by the residual valency of the three
phenyl groups. This conception of a trivalent carbon atom is
really not so extraordinary as it seems ; for we might consider
that ethylene derivatives contain two adjacent carbon atoms of

1 Walden, Ber., 1902, 35, 2018; Gomberg, ibid., 2045. Compare Gomberg,
Ber., 1905, 38, 1342.

2 Gomberg, J., Awer. Chew. Soc., 1900, 22, 757; Ber., 1900, 33, 3150.

226 RECENT ADVANCES IN ORGANIC CHEMISTRY

this type, instead of writing their structural formulae as we
usually do with a double bond between the two unsaturated
carbons.

In favour of this constitutional formula for triphenyl-
methyl we may urge the evidence derived from the reactions of
the substance with iodine and with oxygen, both of which can
be expressed quite simply : —

(C6H5)3C I (C6H5)3C . I

(C6H5)3C I " (C6H5)3C.I

(C6H5)3C 0 (C6H5)3C-0

+ II = I

(C6H5)3C 0 (C6H5)3C-0

And we might also adduce the simplicity of the formulae for
the double compounds of triphenylmethyl with ethers, ketones,
nitriles, etc.

All that this amounts to, however, is that we can express
these reactions in a straightforward manner on the assumption
of trivalent carbon. If we can express them equally con-
vincingly by means of a formula containing only quadrivalent
atoms, then we should be entitled to reject the trivalent
carbon view as adding an unnecessary assumption to our
usual ones.

But there are facts which do not agree with the trivalent
carbon view. In the first place, Gomberg has shown that in
* solution triphenyl methyl has a molecular weight corresponding
. to a formula —

2[(C6H5)3C]

that is to say, the molecular weight is that of hexaphenyl-
ethane or some isomer of that substance. Secondly, G-oinberg
and Cone l have shown that the three phenyl radicles do not
possess identical properties, as they should do if the substance
actually had the triphenylmethyl structure. We need only
outline their proof here, as we shall have to return to it in a
later section. By subjecting para-rosaniline to Sandmeyer's
reaction they obtained tri-p-bromo-triphenyl carbinol, which,
by the action of hydrochloric acid, was transformed into tri-p-
bromo-triphenylmethyl chloride : —

1 Gomberg and Cone, Ber., 1906, 39, 3274.

THE TRIPHENYLMETHYL QUESTION

227

— C— 01

When this substance was treated in the usual way with silver,
it gave a substance analogous to triphenylmethyl. This new
compound formed a peroxide just as triphenylmethyl does, and
therefore (if the trivalent carbon idea be correct) we may safely
assume that it is tri-p-bromo-triphenyl-methyl : —

Now the tri-p-bromo-triphenyl chloride was sealed up in an
air-free flask with excess of molecular silver, and the whole
was shaken for a considerable time. At the end of this, it was
found that the silver had removed all the chlorine (reaction of
triphenylmethyl formation), but in addition it had abstracted one
atom of bromine from the ring of one of the phenyl groups.
Since there was excess of silver present, if all the three phenyl
radicles had identical properties we should expect that they

228 RECENT ADVANCES IN ORGANIC CHEMISTRY

would yield up their bromine simultaneously. Further, the
new compound produced by the elimination of bromine was
not a peroxide similar to that formed by triphenylmethyl, nor
did it yield such a peroxide when exposed to air. The experi-
ments were repeated with other halogen derivatives of tri-
phenylmethyl, and led in these cases to similar results. It is
thus shown : (1) That the substitution of three bromine atoms
in the position para to the " trivalent " carbon of triphenyl-
methyl in no way interferes with the activity of the sutstance ;
(2) further action of silver eliminates only one of the three
bromine atoms, so that one nucleus differs from the other two.
From (1) the complete analogy between triphenylmethyl and
its tribromo-derivative is clear ; and hence we are entitled to
draw the conclusion that the inference in (2) is valid also for
the parent hydrocarbon. But if in triphenylmethyl we have
one phenyl nucleus endowed with properties not shared by the
other two, it is evident that a symmetrical formula —

s — C

cannot give a true representation of the substance's properties.
The triphenylmethyl structural formula with trivalent carbon
has therefore been abandoned at the present day.

3. The Hexaphenyl-ethane Hypothesis.

When Gomberg's hydrocarbon was first prepared, its pro-
perties were found to be so different from what had been
expected of hexaphenyl-ethane that the latter structure was at
once dismissed as incapable of giving a proper representation
of the reactions of the new substance ; but as time went on,
and more information with regard to the properties of the more
highly phenylated ethanes was acquired, it seemed as if the
earlier view had been rather hasty, and that there was a certain
amount of probability in the idea that Gomberg's compound
was, after all, merely hexaphenyl-ethane.

For two years, however, this view was kept in abeyance,
owing to the fact that Ullmann and Borsum L had synthesized
1 Ullmann and Borsum, Ber., 1902, 35, 2877 ; Gomberg, ibid,, 39H,

THE TRIPHENYLMETHYL QUESTION 229

a substance which they regarded as hexaphenyl-ethane. This
body was obtained by reducing triphenyl carbinol; and its
properties corresponded to some extent with those which had
been anticipated for hexaphenyl-ethane. In 1904, however,
Tschitschibabin * established the constitution of this supposed
hexaphenyl-ethane, proving it to be a compound of the follow-
ing structure- : —

x /Cells

CeHg — C — CsHi — CEL

Celv

This removal of the supposed hexaphenyl-ethane from the
literature thus left the possibility open that Gomberg's tri-
phenylmethyl really had the hexaphenyl-ethane structure ; and
Tschitschibabin2 put this suggestion forward, basing his views
on the following considerations.

In the first place, we have to account for the reactivity of
triphenylmethyl, and show why a compound of the hexaphenyl-
ethane structure should be reactive. Tschitschibabin pointed
out that an accumulation of electro-negative atoms or radicles
in a molecule tended to make it much less stable. We have
already seen an example of this in connection with Zincke's
work on the chlorination of phenols ; the accumulation of
chlorine atoms in the compound leads to its degradation into
simpler substances. Again, spacial factors sometimes come
into play and cause a saturated substance like trimethylene to
behave as if it were an unsaturated hydrocarbon. These con-
siderations show that we must be prepared for certain anomalies
and must beware of judging problems of constitution on too
rigid lines. Further, it is not necessary to assume an unsatu-
rated structure for triphenylmethyl merely in order to account
for its ready reaction with :oxygen to form a peroxide, for
Gomberg 3 himself has shown that the fully saturated analogue
triphenyl-iodo-methane reacts in a similar manner. A further
point in favour of the hexaphenyl-ethane view is Gomberg's
proof that in solution his triphenylmethyl had a molecular
weight agreeing with the hexaphenyl-ethane formula rather
/ than with that of tri-phenylmethyl. Nor is this all; for

1 Tschitschibabin, Ber., 1904, 87, 4709. 2 Ibid.

3 Gomberg, Ber., 1902, 35, 1836.

230 RECENT ADVANCES IN ORGANIC CHEMISTRY

when we examine more carefully the behaviour of the highly
phenylated ethane derivatives we shall find that they are
by no means so stable as analogy would lead us to expect.
Tschitschibabin l has proved that even below its melting-point
pentaphenyl-ethane is attacked by air; at a temperature of
only 150° C. hydrochloric acid in benzene solution acts on it
so powerfully that the bond between the two ethane carbon
atoms is broken, and such products as tetraphenyl-ethane, tri-
phenylmethane, and triphenyl-chloro-m ethane, are formed;
while Cone and Kobinson 2 found that the action of phosphorus
pentachloride in boiling benzene broke down the pentaphenyl
derivative into triphenylmethyl chloride.

Against the hexaphenyl-ethane hypothesis we may adduce
several arguments. In the first place, triphenylmethyl is a
colourless solid, but its solutions are deep yellow in tint : no
ordinary benzenoid derivative is known which behaves in this
way. Stronger evidence is to be found in the work of Gomberg,
which we mentioned in the previous section, by which he
showed that one phenyl group had properties different from
those of the others. The ordinary hexaphenyl-ethane formula
gives no indication of this. Thirdly, Gomberg3 has proved
that his hydrocarbon can easily be converted into that which
was obtained by Ullmann and Borsum. On the hexaphenyl-
ethane hypothesis, this reaction would take the following
course, which is parallel to that which is taken in the
semidine change —

-- > (C6H5)3C . C6H4 . CH

C6H4.H C6H5

K . C6H4 . NH . NH . C6H4 . H > E . C6H4 . NH . C6H4. NH2

t I

But Jacobson,4 the greatest authority on the benzidine and
semidine changes, regards such a change in the triphenylmethyl

1 Tschitschibabin, Ber., 1907, 40, 367.

2 Cone and Kobinson, Ber., 1907, 40, 2160.

• Gomberg, Ber., 1902, 35, 3918 ; 1903, 36, 376.
4 Jacobson, Ber., 1904, 37, 196.

THE TRIPHENYLMETHYL QUESTION 231

series as most unlikely. Lastly, we have already seen that
one of the most marked characteristics of triphenyl methyl
is its capacity for forming double compounds with solvents ;
but no such property seems to be possessed by compounds
analagous to hexaphenyl-ethane.

From the foregoing paragraphs, it is clear that the arguments
both in favour of and against the hexaphenyl-ethane view
depend to some extent upon analogy ; and we must be careful
not to lay too much stress upon them unless we are satisfied
that the analogies really hold good. If we rule out the
arguments based upon what a compound " ought " to do, it will
be seen that the evidence remaining — Gomberg's differentiation
between the phenyl nuclei — tells against the hexaphenyl-ethane
hypothesis.

4. Quinonoid Hypotheses.

If we reject the two hypotheses which we have dealt with
in the preceding sections, it is clear that we have still a third
possibility open to us ; for both the triphenylmethyl view and
the hexaphenyl-ethane explanation were based on the assump-
tion that the phenyl nuclei in triphenylmethyl were benzenoid
in character, so that by assuming a quinonoid structure for the
substance we shall arrive at totally different types of formulae.
The quinonoid conception of triphenylmethyl was put forward
very early in the compound's history by Kehrmann.1

This suggestion, involving as it does the assumption of a

1 Kehrmann, Ber., 1901, 84, 3818; see also Norris and Sanders, Am. Chem.
J., 1901, 25, 117; and Gomberg, Ber., 1902, 35, 1824.

232 RECENT ADVANCES IN ORGANIC CHEMISTRY

divalent carbon atom, meets with little approval at the present
time ; and since other formulae of the quinonoid type have since
been suggested which do not necessitate such a postulate, we
need not deal further with this one.

In 1903, Heintschel 1 proposed the formula below —

C6H5x /CH-CH, /CH=CHX /Cells

)C=C( )CH— CH( )C=CC

C6H/ XCH=CH/ XCH=CIF XC6H5

On this hypothesis, the first step in the synthesis of triphenyl-
methyl is the conversion of triphenyl-chloro-methane into a
desmotropic form in which the chlorine atom has been shifted
into a position para to the methane carbon atom —

C6H5\ /CH=CIL C6H5v /CH=CHX /H

< XCH -> xc=c xc

C6H| CH— CH C6H5 CH-CH C1

By the action of metals, two chlorine atoms are withdrawn from
two molecules of the chloro-compound, and in this way
triphenylmethyl is produced —

Civ /CH— CHL

xc o

XC6H5

- 2AgCl

C6H5\ /CH=CHX /CH-CHV /C6H5

XC=C XCH— CH

C6H5

An examination of Heintschel' s formula will show that it
contains two quinonoid phenyl nuclei, Jacobson 2 proposed to
modify this, making only one phenyl group quinonoid, as shown
below —

C6H5 H CH-CH C6H5

\ \/ \ /

C6H5— C— C C=C

C6H5 CH-CH C6H5

1 Heintscliel, Ber., 1903 ; 36, 320, 579.

2 JacobeoD, Ber., 1901, 37,196.

THE TRIPHENYLMETHYL QUESTION 233

This view makes triphenylmethyl a derivative of a substance
approaching the quinole type ; and as we have already seen
that the reactivity of the quinoles is quite abnormal, we might
expect considerable reactive power from a body of the structure
proposed by Jacobson. The change of the Gomberg hydrocarbon
into the substance prepared by Ullmann and Borsum can also be
easily explained on this hypothesis, as the wandering of a single
hydrogen atom is sufficient to account for the isomerization —

OeHs H ': CgHs

\ I /f 1 TT \

C6H5— C— / "\=C 5->C6H5— C-

The Jacobson formula helps us to understand the fact that this
substance, containing six phenyl radicles, can act as if it had
the constitution of triphenylmethyl ; for if it be assumed that
the molecule is decomposed by halogens in such a way that the
single bond between the quinonoid nucleus and the adjacent
carbon atom is loosened, then we should have two " triphenyl-
methyl " radicles set free which would at once react with halogen
atoms giving two molecules of triphenylmethyl halide. The
quinonoid formula also makes clear the meaning of the experi-
ments of Gomberg and Cone 1 to which we made reference in a
previous section. Let us take for example the case of tri-p-
bromo-triphenylmethyl chloride —

-C— Cl

Gomberg and Cone,^er., 1906, 39, 3274.

234 RECENT ADVANCES IN ORGANIC CHEMISTRY

it is clear that, when it is converted into triphenylmethyl by
the action of metals, one of the phenyl radicles must become
quinonoid ; and an examination of the formula of the substance
which would be formed if the quinonoid view be correct will
show that one of the halogen atoms (marked with an asterisk)
should possess the properties of a halogen atom attached to an
aliphatic chain rather than those which are shown by halogen
atoms bound to aromatic nuclei —

*Br

Now, such a halogen atom will be more easily attacked by
metals than will be the case with the other bromine atoms in
the compound in question ; so that we should expect that the
action of an excess of, say, silver upon the tri-^-bromo-triphenyl-
methane chloride will result in two reactions, the first of which
will lead to the elimination of two chlorine atoms, giving rise
to the compound whose formula is shown above, while the
further action of the silver will remove two bromine atoms
from two molecules of this body, the result being the formation
of a substance having the constitution shown below —

(CeEUBr^C^ V7 V" "Vc(C6H4Br)J

(C6H,Br)3CX \— /

— Bro

(C6H4Br)2C=

C(C6H4Br)3 (C6H4Br3C

The results obtained experimentally by Gornberg and Cone
proved that one of the phenyl radicles did actually change from
the benzenoid to the quinonoid form ; but in the view of these
experimenters the assumption of this change alone was not
sufficient to account fully for the problems which the properties
of triphenylmethyl suggest.

We must now turn to examine the objections which have
been brought against the quinonoid view.

THE TRIPHENYLMETHYL QUESTION 235

Tschitschibabin 1 points out that one of the most speedy
and apparently simple reactions which the triphenylmethyl
derivatives undergo is the formation of the peroxide —
(C6H5)3C— 0-0-C(C6H5)3

but that if we are to explain this according to the Jacobson
formula we should have to assume an extremely complicated
isomeric change as the first step in the process.

Gomberg and Cone 2 draw attention to the fact that Jacobson
makes triphenylmethyl a derivative of a substance analogous
to a secondary quinole —

H H

C(C6H5)3 OH

But since secondary quinoles have not yet been proved to be
capable of existence, these authors consider doubtful the
existence of compounds of the Jacobson type. Furthermore,
if we grant the possibility of their existence, it is probable that
they will behave like ordinary quinoles, and hence their reactions
with acids should resemble to some extent the rearrangements
which quinoles undergo under the same conditions. As we
have seen in the chapter upon quinoles, the alkyl group in
these substances usually wanders to the ortho-position ; whence
by analogy the substance produced by the action of acids upon
triphenylmethyl (Ullmann and Borsum's hydrocarbon) should
be represented by the formula (I.) and not by (II.), though
Tschitschibabin believed that (II.) was formed. These arguments,
as the authors themselves admit, are purely theoretical, and
depend largely upon negative evidence.

(C6H5)2CH-<f > (C6H5)2CH-4 >-C(C6H5)3

(I.) (II.)

From a somewhat similar standpoint Auwers 3 has criticized

1 TschitscLibabin, Ber., 1905, 38, 771.

2 Gomberg and Cone, Ber., 1905, 38, 2455.

3 Auwers, Ber., 1907, 40, 2159.

236 RECENT ADVANCES IN ORGANIC CHEMISTRY

the Jacobson formula. He points out that the para-methylene
quinonoid derivatives show such a tendency to revert to the
benzenoid structure that in some cases a profound intramolecular
change may take place. For example, in the compound (I.)
below, the group — CHC12 wanders from its original position to
the atom next the para-carbon atom in order to facilitate the
formation of the benzenoid ring (II.) in preference to the
quinonoid one : —

UH2.CHCla
(I.) (II.)

By analogy, it seems hardly likely that the hydrogen atom
marked with an asterisk in the Jacobson formula would remain
fixed in its present position when by a similar wandering to the
para-carbon atom it could allow the compound to revert to the
benzenoid type.

-CH(C6H5)

That such a wandering must be possible is shown by the con-
version of the Jacobson compound into that of Ullmann and
Borsum by the action of acids; but it seems strange that
a compound of the Jacobson formula should exist in the free
state at all.

Against the Heintschel formula, it has been alleged by
Tschitschibabin * that it should be easily isomerized into a
compound having the structure (B); whereas in practice no
such change takes place.

1 Tschitscliibabin, Ber., 1905, 38, 771.

THE TRIPHENYLMETHYL QUESTION 237

(C6H5)2CH-<f >-<< >-CH(C6H*)2

From the foregoing summary it will be seen that the argu-
ments both in favour of and against the quinonoid structure for
triphenylmethyl are based very largely upon considerations of
what a compound ought to do if it has a structure analogous to
some other compound, the latter body being as yet undiscovered
in practice. As far as the relevant evidence is concerned,
it certainly goes to show that the quinonoid formula is a step in
advance of either the triphenylmethyl hypothesis or the
hexaphenyl-ethane view.

5. The Tautomerism Hypothesis.

We have now exhausted the possibilities of static formulse
to explain the behaviour of triphenylmethyl ; and it is evident
that the results have not been completely satisfactory. All the
three views which we have discussed in the foregoing sections
have certain advantages ; and each has its own drawbacks. It
thus becomes clear that, if we are to make any further progress
towards a solution of the problem, we must contrive some means
of uniting the advantages of the various formulae ; while at the
same time we must endeavour to minimize their weak points.
In order to do this it is obvious that we must turn to modern
dynamic ideas and represent triphenylmethyl as a series of
equilibrium mixtures of isomerides.

Gomberg1 has developed this line of thought; and if his
results do not represent the truth, it seems probable that they
come very close to it. At the present time we cannot assume

* Gomberg, Ber., 1907, 40, 1880.

238 RECENT ADVANCES IN ORGANIC CHEMISTRY

that the problem is completely solved ; but we are evidently
very near to the solution.

G-omberg's later views took their rise in the fact that there
are two varieties of triphenylmethyl which differ from each
other in colour ; the solid form of the substance is colourless ;
but in solution this is changed into a yellow compound.
Schmidlin1 states that he has obtained the two forms of the
substance in solution. Now, Gomberg assumes in the first place
that there are two tautomeric forms of triphenylmethyl, CasHao ;
and in the second place that the radicle triphenylmethyl,
(C6H5)3C, can exist as such and is also capable of tautomerization.
Let us now take up the possible constitution of the solid,
colourless modification. This we may suppose to be hexaphenyl-
ethane. It is evident that we may assume tautomeric change
in this compound, leading us to the following structure : —

This alteration of the benzenoid into the quinonoid form would
be accompanied by a change of the substance from colourless to
yellow ; and since all ordinary solvents seem to be capable
of yielding yellow solutions of triphenylmethyl, we may assume
that this change from the benzenoid to the quinonoid form
takes place under the action of most solvents during the process
of solution.

We must now go a step further and deal with the behaviour
of triphenylmethyl dissolved in a medium of high dissociating
power, liquid sulphur dioxide. It has been proved by Walden 2
that a solution of the hydrocarbon in this solvent possesses
a fairly high conductivity, and that the molecular conductivity
increases with the dilution ; in other words, the substance
behaves just like an ordinary ionized salt. From this behaviour,
Gomberg deduces that tautomerization is not the only change
which triphenylmethyl undergoes as it is dissolved ; but that
in addition it is dissociated into two ions which we may

1 Schmidlin, Ber., 1908, 41, 2471.

2 Walden, Zeit. Phys. Chem., 1903, 43, 443 ; Gomberg and Cone, Ber., 1904,
37, 2403.

THE TRIPHENYLMETHYL QUESTION 239

represent as below. The anion is supposed to have the
benzenoid structure, while the kation is quinonoid.

H-.+

(C6H5)2C : C6H4^ (Quinonoid)

(Benzenoid)
[(G,H^G-

On this view, the action of iodine upon triphenylmethyl solu-
tions is explicable. The iodine in solution is supposed to
interact with both the anion and the kation, yielding one
molecule of benzenoid triphenylmethyl iodide and one molecule
in the quinonoid form; but since the latter seems to be
incapable of existence in the free state, it is assumed that it
undergoes intramolecular change at once and produces a
benzenoid molecule. When we turn to the action of oxygen
upon triphenylmethyl in solution, however, we have a somewhat
different state of affairs, since only the anion unites with oxygen.
(This follows from the fact that the peroxide formed has the
benzenoid structure, whereas the action of oxygen upon the
quinonoid ion would give rise to a highly complicated product
which is not observed among the reaction products). We are
thus led to the further assumption that in the process of peroxide
formation the first step is the oxidation of the benzenoid ions ;
as these are removed from the solution, equilibrium is disturbed ;
and, in order to re-establish it, some of the quinonoid ions must
re-tautomerize into the benzenoid form. They in turn are
removed by the oxygen ; and the process continues until all the
triphenylmethyl is exhausted.

The same tautomerization process can be invoked to explain
why triphenylmethyl gives a yellow solution with ethers, esters
and ketones, while the solid double compounds which crystal-
lize out from these solutions are colourless. In this case the
benzenoid ions may be assumed to unite with the quadrivalent
oxygen of the ethers, etc. ; and in order to take their place some
of the quinonoid ions are converted into benzenoid ones.

According to Gomberg, then, we can explain all the im-
portant properties of triphenylmethyl on the basis of the

240 RECENT ADVANCES IN ORGANIC CHEMISTRY

following hypotheses : (1) tautomerization of solid triphenyl-
methyl to a quinonoid substance having the Jacobson formula ;
(2) partial dissociation of this compound into positive and
negative ions in all solvents ; (3) mutual interconvertibility of
these ions by tautomeric change. Gomberg further assumes
that solid triphenylmethyl is actually a free radicle, — C(C6H5)3.

6. Conclusion.

We have now examined in turn the principal hypotheses
which have been put forward at various times to account for
the peculiar properties of triphenylmethyl; and it must be
admitted that even yet there appears to be some doubt as to
whether we have arrived at the solution of the problem. The
substance in its general behaviour is so unlike the majority of
organic compounds that at first sight the tri-valent carbon
hypothesis appeared to have some points in its favour ; but, as
we have seen, the arguments against it are strong. It seems
clear that the quinonoid view, however it be expressed, is the
most probable of all ; and, in its final form as put forward by
Gomberg, it seems to present the maximum of advantages. In
certain respects triphenylmethyl appears to resemble the quin-
hydrone type of compound, and it may be that further re-
searches in that branch of the subject may throw light upon
the curious behaviour of this enigmatic body.

CHAPTEE XI

ASYMMETRIC SYNTHESES AND NEW METHODS OF PRODUCING
OPTICALLY ACTIVE COMPOUNDS

IF we pass a current of carbon dioxide into water in which
sticks of magnesium have been immersed, some of the gas is
reduced to formaldehyde l ; and by acting upon the latter with
calcium hydrate solution,2 we can produce fructose —

CH2OH . (CH . OH)3 . CO . CH2OH

A similar operation is probably carried out in the natural
synthesis of fructose in plants, where the starting materials
are also water and carbon dioxide; but the process in the
latter case is evidently subject to some influence which is
absent in our ordinary laboratory reactions, for the naturally
occurring fruit sugar is optically active, while that prepared
in the laboratory has no action upon the plane of polarization.
The question now suggests itself, in what way does the
mechanism of the reaction vary in such a manner as to produce
results so similar up to a certain point, and yet so distinct from
one another ?

It is at once obvious that we cannot consider the organism
of the plant merely as a peculiar kind of beaker in which the
reaction takes place; for if this were so, there would be no
difference between the two reaction products. Evidently, then,
the plant tissues play some part other than merely containing
the interacting substances; they absorb water and carbon
dioxide at their surfaces, bring the two compounds together,
and then in some way assist them to react with one another
so as to form fructose. The accepted view is that the plant
tissues combine with both the carbon dioxide and the water
in the first place to form some unstable substance, and then

1 Fenton, Trans. Chem. Soc., 1907, 91, 687.

2 Loew, J. pr. Chem,, 1886, 33, 321.

242 RECENT ADVANCES IN ORGANIC CHEMISTRY

eliminate fructose as a decomposition product. If we accept
this view, the problem before us becomes clear at once. In
the case of our laboratory reaction, we are dealing throughout
with a purely symmetrical set of substances, and there is no
possibility of optically active products being formed. But in
the case of the plant we have a mass of optically active bodies
which make up the sap and tissues of the organism; these
fasten upon the water and carbon dioxide, combine with them,
forming more complex but still optically active compounds,
which on decomposition eliminate optically active products.

In recent years many attempts, some successful and others
unavailing, have been made to parallel this process by means
of our ordinary laboratory reactions. To do this, we must
take as our starting-point some optically active body which
will play the part of the active constituents in the plant; to
this optically active nucleus we must then add a new chain
of atoms, when we shall have a parallel to the unstable inter-
mediate product in the plant ; and now, when we split off the
new chain from the original active nucleus, we must find that
it in turn is active if our artificial reaction is to resemble the
natural one. Perhaps a concrete example will make the matter
clearer. Let us take an optically active acid, X . COOH, and
combine it with amidobenzaldehyde. We shall then have the
compound —

X.CO.NH.C6H4.CHO

To this we then attach a molecule of hydrocyanic acid to
form the cyanhydrin—

X . CO . NH . C6H4 . CH(OH) . CN

This represents the intermediate unstable product in the
plant; and it will be noticed that a new asymmetric carbon
atom (marked with an asterisk) has been produced in the
compound. On hydrolyzing the nitrile group, we shall obtain
the corresponding acid, in which the new asymmetric carbon
still remains —

X . CO . NH . C6H4 . CH(OH) . COOH

If we now wish to represent the breakdown of the unstable
plant product, we need only split off the original active acid

ASYMMETRIC SYNTHESES 243

group from this, and we shall have left behind the " new chain
of atoms " —

NH2 . C6H4 . CH(OH) . COOH

Should this prove to be active, the parallel between plant
synthesis and laboratory reaction will be complete.

As far back as 18 89,1 evidence bearing indirectly upon this
question had been obtained from researches in the sugar group.
For example, it had been shown that dextro-mannose (I.),
which, when submitted to the cyanhydrin reaction, should
theoretically yield an equirnolecular mixture of the compounds
(II.) and (III.), actually in practice produces only one of the
isomers.

OHO
HO.H
HO.H
H.OH
H.OH
CH2OH \

(I-) \
CHO / \ CHO

HO.H i \ H.OH

HO . H HO . H

H . OH H . OH

CH2OH CH2OH

(II.) (III.)

In this case, therefore, the original asymmetric nucleus has
governed the progress of the reaction to such an extent as to
preclude the formation of one isomer. We are, however,
unable to carry the matter further in this particular example,
as we have no means of splitting off the original mannose
molecule without destroying the " new chain " in the process.

Having now shown what is being sought in this branch of
the subject, it may be well to give at this point a formal defi-
nition of the term " asymmetric synthesis" In an asymmetric
synthesis an active compound is taken as a starting-point, to
which a new radicle is added in such a way as to form a new
asymmetric carbon atom ; the originally active portion of the

1 Fischer and Hirschberger, Ber., 1889, 22, 305; Fischer, Per., 1891, 27,
3208 ; Fischer, Annalcn, 1892, 270, 68.

244 RECENT ADVANCES IN ORGANIC CHEMISTRY

molecule is then split off, and the remainder must be optically
active if the synthesis is successful.

We must next review the various attempts which have
been made to solve this problem. The first of any interest is
that made by Cohen and Whitely,1 in which they took as their
starting material mesaconic acid. This they esterified with
Isevo-menthol, and in this way obtained an active ester. They
next reduced this to methyl-succinic menthyl ester, and finally
split off the menthyl group. If the synthesis had been an
asymmetric one, the remaining methyl-succinic acid should
have been active ; it was not.

It is unnecessary to refer in detail to Fischer and Slimmer's
unsuccessful method,2 in which helicin was the starting-point ;
or to Cohen and Whitely's second unsuccessful attempt to
produce optically active bodies in this way.3 The first true
asymmetric synthesis was carried out by Marckwald 4 in the
following ingenious way. He took methyl-ethyl-malonic acid
(I.), which contains no asymmetric carbon atom, and from it
made the acid brucine salt (II.). When this salt is heated
until no more carbon dioxide is liberated, the brucine salt of
valerianic acid (III.) is left, and it will be noticed that this
acid contains a new asymmetric carbon atom. Marckwald
found that when he liberated the valerianic acid from the salt
(III.) it had a slight optical rotatory power, which could only
be due to the new asymmetric carbon atom.

C2H5 COOH C2H5 COOH C2H5 H

V V V •

/\ /\ /\

CH3 COOH CH3 COO (Brucine.) CH3 COO (Brucine.)
(I.) (II.) (III.)

By this research it was proved that asymmetric synthesis
was not an impossible achievement, and since that time several
such syntheses have been carried out.

McKenzie5 has produced a mandelic ester in which the

Cohen and Whitely, Proc. Chem. Soc., 1901, 16, 226.

Fischer and Slimmer, Sitzungsber. d. K. Akad. Wiss. Berlin, 1902, 597 ;
Ber , 1903, 36, 2575.

Cohen and Whitely, Trans. Chem. Soc., 1901, 79, 1305.

Marckwald, Ber., 1904, 37, 349, 1368, 4696.

McKenzie. Trans. Chem. 8oc.t 1904, 85, 1249 ; 1906, 89, 365.

ASYMMETRIC SYNTHESES 245

Igevo-form predominates over the dextro-form. Choosing as
his starting substance benzoyl-formic acid (L), he esterified this
with laevo-menthol, producing the ester (II.). This he reduced
by means of aluminium amalgam, and thus obtained mandelic
Isevo-menthyl ester (III.) —

C6H5 . CO . COOH C6H5 . CO . COOCi0H19

(I.) (II.)

C6H5— C— COOC10H19

(in.)

On examination, (III.) was found to be a mixture of c2-man-
delic-^-menthyl ester with /-mandelic-/-menthyl ester; the
latter slightly preponderated in the mixture, so that on getting
rid of the menthol, the asymmetric synthesis would be com-
plete. This method has been applied to the preparation of
active lactic from pyruvic acid.1

Using benzoyl-formic menthyl ester again as his starting-
point, the same author 2 applied the Grignard reaction, with
the following result : —

OMg.I OH

C6H5.CO.COOC10H19 C6H5.C.COQC10H19 C6H5.C.COOC10H19

! I

CH3 CH3

(I.) (II.) (III.)

From the menthyl ester (I.) the compound (II.) was obtained
by the action of magnesium methyl iodide ; this intermediate
compound was then decomposed with water to form (III.);
from which in turn, by the action of acid, a mixture of exter-
nally compensated and Isevo-methyl-phenyl-glycollic acid was
produced. Since the one antipode predominated over the other,
the asymmetric synthesis had been successfully accomplished.
McKenzie and Wren3 have been able to synthesize both

1 McKenzie, Trans. Chem. Soc., 1905, 87, 1373 ; McKenzie and Wren, ibid.,
1906, 89, 688.

2 McKenzie, Trans. Chem. Soc., 1904, 85, 1249.

3 McKenzie and Wren, Trans. Chem. Soc., 1907, 91, 1215.

246 RECENT ADVANCES IN ORGANIC CHEMISTRY

laevo and dextro forms of tartaric acid in the following manner.
Fumaric acid was esterified with Isevo-borneol, the ester was
dissolved in glacial acetic acid and oxidized with potassium
permanganate. The result was a tartaric ester in which there
was a slight excess of the Isevo form.

CH . COOH CH . COOCioHn CH(OH) . COOCi0Hi7

I! II I

CH • COOH CH • COOC10H17 CH(OH) . COOC10H17

By using dextro-borneol they obtained the dextro-tartaric-
dextro-borneol ester in excess over the laevo-tartaric-dextro-
borneol ester. When menthol was substituted for borneol, it
gave a greater yield of the Isevo-tartaric ester.

Only two other attempts in this direction need be dealt
with here. In the nitrogen series, Scholtz 1 has endeavoured to
carry out an asymmetric synthesis by the action of optically
active halogen alkyl derivatives upon racemic bases ; E. and 0.
Wedekind 2 have also made experiments of the same type, but
in neither case was a true asymmetric synthesis accomplished.

Smiles 8 investigated a case of asymmetric synthesis of
tetravalent sulphur, but found that when methyl-ethyl-sulphide
interacted with laevo-menthyl-bromacetate an ester (I.) was
produced which, on hydrolysis with hydrochloric acid, gives the '
externally compensated thetine (II.). The two antipodes are
therefore formed in equal quantity, and no asymmetric synthesis
takes place.

CH3 Br CH3 O

\/ \/\

8 S CO

C2H5 CH2.COOC10H19 C2H5 CH2

(I.) (ID

We must now leave the question of asymmetric synthesis
and consider some other recently discovered methods of
producing optically active compounds.

In his study of the tartaric acids, with which modern
stereochemistry may be said to have begun, Pasteur contrived
three methods by which optically active substances could be

1 Sclioltz, J5er., 1901, 34, 3015.

2 E. and O. Wedekind, Ber., 1908, 41, 456.

3 Smiles, Trans. Chem. 8oc., 1905, 87, 450.

ASYMMETRIC SYNTHESES 247

obtained from externally compensated mixtures. One method
depended upon the spontaneous separation of the crystals of the
two antipodes ; another upon the selective action of fungi ; and
the third upon the formation of salts with an active substance
and the racemic compound as the two components. No greater
tribute to Pasteur's genius can be found than the fact that for
nearly fifty years, in spite of almost incessant research in this
field, no substitutes for these three methods had been invented.
Modifications may have been introduced, such as substituting
hydrazone formation for salt formation,1 but no fundamentally
different method was contrived until a few years ago.

We may now examine in turn the methods which recent
workers have brought forward. As regards the question of
spontaneous separation of the antipodes, no advance seems
possible. The study of transition temperatures has, of course,
rendered the use of this mode of separation much more certain
in its results than it used to be, but in actual experimental
details it remains as Pasteur left it. Numerous attempts have
been made2 to modify it by using optically active solvents
instead of symmetrical ones, but none of these have been
successful ; and it appears improbable that resolution can be
accomplished in this manner. It has been shown,3 however,
that if a saturated solution of a cW- compound be sown with
crystals isomorphous with those of one antipode, that antipode
will predominate in the crystals formed. The seed crystal need
not necessarily be itself optically active.

Turning now to the biochemical method of separation by
the aid of fungi, though no modification of this has been
introduced which would enable us to obtain one component of a
racemic substance by a less wasteful method than the original
Pasteur one, Bertrand4 has applied the sorbose bacterium

1 Erlenmeyer, junr., Her., 1903, 36, 976; Erlenmeyer, junr., and Arnold,
Annalen, 1904, 337, 307 ; Neuberg, Ber., 1903, 36, 1192 ; Neuberg and Federer,
Her., 1905, 38, 801.

2 Tolloczko, Zeit. pliyg. Chem., 1896, 20, 412; Cooper, ibid., 1898, 26, 711 ;
Amer. Chem. J., 1900, 23, 253 ; Kipping and Pope, Proc. Chtm. Soc., 1898, 14, 113 ;
cf. Wedekind, Ber., 1908, 41, 457, footnote ; Jones, Proc. Cam. Phil Soc., 1907
14, 27.

3 Ostromisslensky, Ber., 1908, 41, 3035 ; Kipping and Pope, Trans. Chem.
Soc., 1909, 95, 103.

4 Bertrand, C. r., 1896, 122, 900; 1898,' 126, 762; Bull. soc. chim., 1898, III.
19, 347, 947, 999.

248 RECENT ADVANCES IN ORGANIC CHEMISTRY

(Bacterium xylinum) in a somewhat different manner. This
bacterium has the faculty of oxidizing certain hydroxyl-
compounds, attacking secondary hydroxyl groups which lie in
the a-position to a primary hydroxyl, and transforming them
into carbonyl groups. It is found, however, that the configura-
tion of the sugar submitted to the bacterium has a considerable
effect upon the action, for only those sugars are attacked which
have the group (I,), those having the group (II.) at the end of
the chain being uninfluenced by the ferment. It will be seen
that in (I.) the two hydroxyl groups are adjacent, while in (II.)
a hydrogen atom and a hydroxyl group lie together.

OH OH OH H

_C— C— CH2OH — C — C— CH2OH

II II

H H H OH

(I.) (II.)

For example, xylite is unattacked by the sorbose bacterium,
since it does not contain the grouping (I.) —

H OHH

CH2OH— C— C— C— CH2OH

I I I
OHH OH

while arabite, which contains the grouping (II.) is oxidized to
the corresponding keto- compound —

OH OH H OH H

-C— C— C— CH2OH->CHoOH— CO— C-

II II

H OH

CH2OH-C— C— C— CH2OH->CH2OH— CO— C— C— CH2OH

I I I
H H OH

Coming now to chemical methods of separating one
antipode from another, we shall find that in the last few years
a very considerable number of new lines have been struck out
in this branch of the subject.

The Pasteur method of separating antipodes by means of
the salt formation was in its essence a purely static one. In
order to resolve a racemic base into its antipodes, a salt was
made in whose preparation an active acid was used. The two
salts would then be, say, cZ-acid-d-base and d-acid-/-base ; and

ASYMMETRIC SYNTHESES 249

since they were now no longer optical antipodes, they could be
separated from one another by utilizing their differences in
solubility. This was a purely physical method, in which the
chemical reaction played, per se, no part in the actual sifting of
one compound from the other.

Fischer1 attacked the matter from quite a different stand-
point. For him, the chemical reaction, instead of being a
subsidiary part of the separation, became the actual machinery
of resolution. Taking the case of the hydrolysis of cane-sugar,
he applied an asymmetric hydrolyzer, and hoped in this way
that a selective hydrolysis might be achieved. Unfortunately,
his experiments with dextro- and Isevo-camphoric acids, as well
as others with camphor-sulphonic acids,2 were alike unsuccess-
ful ; no selective hydrolysis could be observed, both agents
hydrolyzing the sugar with the same velocity. More interesting
results were obtained by Fischer3 in studying the action of
enzymes on glucosides. Selective hydrolysis takes place in
this case, and thus a separation of isomers is possible.

A further step in this direction was taken by Marckwald
and McKenzie.4 Since the reaction of salt-formation is ionic,
it takes place almost instantaneously, and spacial influences
appear to have very little bearing upon it. On the other hand,
a comparatively slow reaction, like esterification, which seems
to require the formation of an unstable intermediate product,
should lend itself better to selective action. Marckwald and
McKenzie proceeded on this assumption, and carried out a
series of experiments in which they esterified a racemic acid
with an active alcohol, interrupting the process before all the
acid was esterified. In this way, if one antipode reacted more
rapidly than the other with the alcohol, an excess of its ester
would be formed. This was found actually to be the case
Lsevo-menthol reacts more readily with dextro- than with
laevo-mandelic acid, so that if the esterification process be
interrupted when half the acid is esterified the ester will
contain some Z-mandelic-Z-menthyl ester mixed with an excess
of d-mandelic-/-menthyl ester. By saponifying the mixed esters
and repeating the process several times one antipode could

1 Fischer, Zeit physiol Chem., 1898, 26, 83.

2 Caldwell, Proc. Roy.8oc.,19Ql, 74, 184.

3 Fischer, Zeit. physiol. Chem., 1898, 26, 83.

4 Marckwald and McKenzie, Per., 1899, 32, 2130.

250 RECENT ADVANCES IN ORGANIC CHEMISTRY

be separated completely from the other. The same authors 1
applied the inverse method of completely esterifying the
racemic mandelic acid with active menthol and then fraction-
ally hydrolyzing the mixed esters with insufficient alkali.
The ester of the type d-acid-/-menthol is hydrolyzed at a
different rate from the type Z-acid-Z-menthol, so that the reaction
product contains an excess of one acid over the other. By
removing the unhydrolyzed esters the mixture of acids is left ;
and this was found to be optically active.

Marckwald and Meth2 have applied a similar method to the
case of amide formation. They have shown that if, when a
racemic acid is converted into an amide by heating it with an
active amine, the process be interrupted before all the acid
is converted into amide the unconverted acid is optically active.
Eacemic mandelic acid, for example, when heated with laevo-
menthylamine to 160° or 170° for ten hours, yields about
36 per cent, of amide. The remaining 64 per cent, of the acid
had a specific rotation of [a]D = - 5'1°. It appears that
the relative velocities of amide formation from Isevo-menthyl-
amine with dextro- and laevo-mandelic acids are in the ratio
1 : 0-682.

Marckwald and Paul 3 have utilized another property of non-
antipodic isomers. They take a racemic acid, form a salt with
an active base in the usual manner, and then heat the salt to a
high temperature for some time. Now, while two optical
antipodes have exactly equal velocities of racemization, two
salts of the types c?-acid-e?-base and /-acid-^-base will racemize
at different speeds; so that when the reaction product is
isolated and the acids regenerated they will be found to be
optically active. In many cases, however, this method yields
no results, doubtless owing to the slight difference between the
racemization velocities of the two isomers.

We may now turn to some other attempts which have
been made to obtain optically active substances from inactive
compounds. We have already seen that it has not been found
possible to separate one antipode by crystallization from an
optically active solvent. There still remains the possibility

1 See also McKenzie and Thompson, Trans. Chem. Soc., 1905, 87, 1001.

2 Marckwald and Meth, Ber., 1905, 38, 801.

3 Marckwald and Paul, Ber., 1905, 38, 810 ; 1906, 39, 3C54.

ASYMMETRIC SYNTHESES 251

that if we start with an inactive substance containing no
asymmetric carbon atom and so modify it, as in asymmetric
synthesis, that an asymmetric carbon atom is formed in it, and
if, further, we carry out the production of this new asymmetric
carbon atom in an active solvent, we might find that the solvent
had had an influence upon the reaction and caused the one
antipode to be formed in excess of the other. Kipping 1 has
put this idea to the test in the following way. He reduced
pyruvic acid to lactic acid in a strong glucose solution ; and
also synthesized benzoin from benzaldehyde and potassium
cyanide, using as solvent a solution of camphor in alcohol.
In neither case was the reaction product active.

Hitherto we have dealt exclusively with the chemical side
of the problem, but it may not be amiss to say a few words
on some experiments which have treated the matter from a
physical point of view, and have led in some cases to results of
considerable interest.

Pasteur appears to have been the first to attempt to produce
optically active substances by use of a strong magnetic field.
His experiments were unsuccessful, as were those of Boyd.2
It appears that an ordinary magnetic field is not truly asym-
metrical ; and in order to introduce the required asymmetry
some addition must be made to the play of forces involved.
Meyer3 contrived the following system in order to overcome
this difficulty. In his apparatus, a magnetic field is obtained
in the usual way, and through it is passed a ray of light. When
this ray of light is polarized before being passed through the field,
the system becomes asymmetrical. Meyer placed a glass beaker
in the path of the ray, and in the beaker he reduced benzoyl-
formic to mandelic acid. The results, however, were negative ;
the mandelic acid formed being the ordinary externally com-
pensated variety. Though this experiment failed, the method
appears to be based upon sound principles, for Cotton 4 has
shown that ^-circularly polarized light is differently absorbed
by the dextro- and Isevo -forms of tartaric acid. Absorption
cannot be carried on without a loss of energy, so that obviously a

1 Kipping, Proc. Chem. Soo., 1901, 16, 226 ; E. and O. Wedekind, Per., 1908,
41, 456.

2 Boyd, " Dissertation;" Heidelberg, 1896.

3 Meyer, Chem. Zeit., 1904, 28, 41.

4 Cotton, Ann. Ghim. Phys., 1896, VII., 8, 373.

252 RECENT ADVANCES IN ORGANIC CHEMISTRY

reaction, carried out under polarized light, favours the forma-
tion of one antipode rather than the other. Henle and Haakh l
have pointed out that results could be expected from such a
method only if the reaction producing the asymmetric carbon
atom were one which is influenced by light action ; but their
experiments in this line were unsuccessful. Unfortunately,
the only apparatus at present at the command of the chemist
does not allow us to influence reactions in this manner ; and
until we have more powerful instruments at our disposal it is
unlikely that syntheses in this way will yield any great results.
In conclusion, we may sum up the matter simply. The
task of producing a laboratory parallel to plant syntheses has
been accomplished; we may not have actually produced the
same substances as the plant forms within its organism, but we
have certainly utilized analogous methods, and obtained similar
results.

1 Henle and Haakh, Ber., 1908, 41, 4261.

CHAPTER XII

SOME THEOKIES OF ADDITION KEACTIONS

WHEN two compounds interact with one another there are two
possible courses : in the one case from the two original mole-
cules we may have two or more new molecules formed ; while
in the second case the two molecules coalesce to form a single
substance. The latter type of interaction is what we understand
by the term "addition reaction."

The importance of addition reactions from the point of view
of theory has been very widely recognized, as is testified by the
flood of hypotheses which have been put forward in this branch
at one time or another. We cannot attempt to give a complete
sketch of the various views which have been suggested, and
must content ourselves with brief accounts of several recent
attempts to formulate the principles underlying the practical
side of the subject. The relation between stereochemical in-
fluences and the products of addition reactions lies outside the
province of the present volume; and we shall confine our
attention to the purely structural side of the question.

There is very little connection between the views which
different authors have brought forward to deal with the addition
question, and consequently it is difficult to arrange the various
theories in anything resembling logical sequence. The most
simple arrangement seems to be to begin with the more general
views, and deal later with those of more restricted scope ; and
this plan will be followed in the rest of the chapter.

The most general view of all has been taken by Michael.1
According to him, addition is caused by the affinity of the two
interacting molecules for each other, and takes place in the
manner which produces the most " chemically saturated " com-
pound. But, as he points out, this " neutralization " of affinity

1 Michael, /. pr. Chem., 1888, II. 37, 524; 1899, 60, 286, 409; 1903, 68,
487; Ber., 1906, 39,2138.

254 RECENT ADVANCES IN ORGANIC CHEMISTRY

depends very largely upon the character of the atoms forming
the two interacting molecules. For instance, suppose that the
two molecules —

A— B C— D

are capable of reacting together. Let us first consider the
affinity of A for C and for D. If the affinity of A for C is
greater than that of A for D, then we should expect to find A
attaching itself to C, and leaving B to attach itself to D, thus
forming the compound —

A B

But this leaves out of account the attraction of B for C. If
this were greater than the attraction of B for D, then we should
expect the formation of the compound —

B A

I I
C— D

It is obvious that the actual result of the reaction will
depend upon the relative intensity of the forces between
A and C and between B and C, coupled with the relative
intensity of the forces between A and D and between B and D.
Let us represent the forces between A and C by ac and those
between B and C by lc, also those between A and D by ad and
those between B and D by Id. Then the forces which are
favouring the formation of the first type —

A B

I I
C— D

(I.)

will be represented by ac + Id, while those favouring the
formation of —

B A

I i
C— D

(II.)

will be represented by be -f ad. The amounts of the two com-
pounds formed during the progress of the reaction will therefore

SOME THEORIES OF ADDITION REACTIONS 255

be to one another in the ratio of ac 4- "bd : lc + ad. This is
termed Michael's " Distribution Principle." —

The consideration of a concrete case will make the matter
clearer. Suppose we take propylene, CH3 . CH : CH2, and allow
it to react with hydriodic acid. Two possible products may
result ; for in the one case the iodine atom may attach itself to
the middle carbon atom, while in the other it may be attracted
by the end carbon atom.

(I.) (IT.)

CHa.CHI.CHa CH3 . CH2 . CH2I (300:1)

It has been found that about three hundred times more of (I.)
is formed than (II.). This is due to the great chemical difference
between the hydrogen and iodine atoms of hydriodic acid. But
if we lessen this difference between the two atoms by substituting
a chlorine atom for hydrogen (using iodine chloride instead of
hydriodic acid), we shall find that the iodine atom now attaches
itself to the end of the chain rather than to the central atom, the
amounts of the compound (la.) and (Ha.) being formed in the
ratio of three to one —

(la.) (lla.)

CH3 . CHOI . CH2I CH3 . CHI . CH2C1 (3 : 1)

The directing influences at work may be still further neutra-
lized if we employ bromine chloride instead of iodine chloride.
In this last case there is great similarity between the two atoms,
and, as a result, the two possible end-products are formed in
very nearly equal proportions (1.4:1.0) —

(16.) (IK.)

CH3 , CHOI . CH2Br CH3 . CHBr . CH2C1 (1.4 : 1)

It will be seen that the " Markownikoff Eule " 1 is only a
particular application of the " Distribution Principle," and that
its applicability depends to some extent upon the constitution
of the molecule containing the double bond.-

"We must now turn to the views of Vorlander.2 If we take
an a/3-unsaturated ketone and allow it to react with an acid,
the first substance formed is a coloured, unstable substance.

1 See p. 58, footnote.

2 Vorlander and Mumme, Ber., 1903, 36, 1470; Vorlandcr and Hayakawa,
ibid., 3528; Vorlaiider, Annakn, 1903, 341, 1 ; 1906, 345, 155.

256 RECENT ADVANCES IN ORGANIC CHEMISTRY

This series of bodies Vorlander designated as Type ^t. These
compounds may undergo a change, being converted into white
stable substances which Vorlander classes together under the
heading Type <$J. According to Vorlander, these two series of
substances differ from each other in the following way.

If we consider different states of unsaturation as the outcome
of the different capacities of the elements and radicles for
positive or negative electricity (of course this is merely used as
an analogy), then in the case of the addition of two oppositely
charged compounds there are two possibilities —

A. There may be no complete neutralization of the electric
charges. In this case the electricity which was originally
spread over the surface of the substances becomes concentrated
into a comparatively small area by the attraction of the opposite
charge on the second substance, and consequently strain is
set up.

B. The compounds unite together and become discharged.
The figures below illustrate this graphically.

Let us take the case of the addition of an acid HX to the
ketone K.CH:CH. CO.R as an example. In compounds of
Class & the acid HX has simply attached itself to the ketone
to form a kind of double compound, which he represents
thus —

(HX)
K.CH CH.CO.K

Here no separation of HX into H and X occurs. But in
the case of the stable bodies, Type 33, this dissociation of the
acid molecule does actually take place, so that we may write
the formula of these bodies thus —

SOME THEORIES OF ADDITION REACTIONS 257

X H H X
K . CH- -CH . CO . E or E . CH CH . CO. E

Now, in order that any addition- compound be formed there
must exist between the two interacting molecules of acid and
ketone a difference which Vorlander terms a difference in
" potential." The amount of this difference we can estimate by
measuring the velocity with which the two substances unite ;
for the greater the difference in potential between them, the
more rapidly will they unite with each other. The rapidity of
formation of the coloured, unstable products is almost instan-
taneous, while the rate of formation of the stable type is quite
measurable. From this we may conclude that the strain in the
case of the formation of the unstable bodies is greater than that
in the production of the stable isomers.

Vorlander expresses his view somewhat as follows. If we
consider the case of two substances about to interact, the
difference in potential between them may be called Ji. When
the unstable compound is formed, only a very little energy is
used up, and the difference in potential between the two
components sinks to ha, the rest of the original energy being
utilized in holding the two components loosely together. Since
there is little change of potential throughout the system, such
reactions can take place rapidly even at low temperatures. In
the case of the stable compounds, however, the difference in
potential h is much reduced, say to hb. On Vorlander's view,
time is required to bring about this change in potential, and
also to overcome certain reaction difficulties, so that the rate of
addition is slow. Further, the two types of addition products,
owing to the difference in potential between them, have quite
different properties. Vorlander groups the whole series of
addition reactions according to their results, and in this way
obtains the following series : (1) Compounds of Type gl ; (2)
double salts ; (3) complex salts ; (4) compounds of Type 23.

The difference between the two systems gl and 23 is
especially marked, when we take into consideration the nature
of the solvent. In the case of gl the dielectric constant of the
solvent will exercise a very marked influence ; but in the case
of 23, since the compounds are in actual contact, the solvent will
have no effect.

In the foregoing theories the question of addition was treated

258 RECENT ADVANCES IN ORGANIC CHEMISTRY

from a broad standpoint, but now we must come to more
restricted fields. It is very seldom that any theory is accepted
immediately after being published ; usually a considerable time
is required during which the chemical world assimilates the
author's views in a more or less unconscious manner, until some
day they find their way into text-books. It is a remarkable
tribute to the value of Thiele's theory, with which we are
about to deal, that it became a classic almost as soon as it was
published ; and was not forced to undergo the process of cud-
chewing which is usually the most that can be expected when
a new theory is under consideration.

The Thiele theory l is based upon the following assumption.
If we imagine the case of a double bond between two atoms, it
is supposed that the whole of the affinity of the atoms is not
used up, but that in addition to that which is utilized in join-
ing the two atoms together there is a slight excess on each
atom. This slight excess of valency Thiele designates by the
name Partial Valency, and to its presence he attributes
the additive power which unsaturated compounds display.
To represent the partial valencies, Thiele employs a dotted line,
thus —

H E
R.C E.C R.N

II II II

R.C 0 E.N

Now, when we come to the consideration of such a system as
R.CH:CH.CH:CH.R

we find that it shows one peculiar property in connection with
addition reactions. Since it contains two double bonds, it
might be expected to take up four atoms of hydrogen or
bromine at once, or at least to take up two atoms of bromine
or hydrogen at one of the double bonds. In other words, we
should expect to find one molecule of bromine attacking it first
with the formation of the compound —

R. CHBr . CHBr . CH: CH. R

1 Thiele, Annalen, 1899, 306, 87.

SOME THEORIES OF ADDITION REACTIONS 259

to which another bromine molecule might be added, giving the
tetrabromo-compound —

K . CHBr . CHBr . CHBr . CHBr . R

In practice, however, the first molecule of bromine does not
attack either of the double bonds ; it attacks them both at once,
with the formation of the compound —

R. CHBr. CH:CH. CHBr. R

in which both of the original double bonds have disappeared,
giving rise to a new double bond in the centre of the molecule.
If we write out the scheme of partial valencies for the original
substance —

R.CH:CH.CH:CH.R

it is evident that only the two at the ends of the system have
the faculty of attracting bromine, the two middle partial
valencies failing to act. In order to express this behaviour
Thiele writes the formula in the following way, in which the
two central partial valencies are supposed to have neutralized
one another : —

R-CH=CH— CH-CH— R

We can make this behaviour clear by supposing that the
carbon atoms of the chain are charged alternately with positive
and negative forces, the two central atoms will then neutralize
one another, leaving the ends still charged —

+ - + - +

R.CH:CH.CH:CH.R R.CH:CH.CH:CH.R

Such a system Thiele terms a Conjugated Double Bond.

If addition takes place in the case of a conjugated double
bond, obviously the two new atoms will attach themselves at the
ends of the chain in the position indicated by the free partial
valencies. But this does not end the matter, for no sooner has
addition taken place than the conjugation is destroyed, and
hence a new double bond will be formed between the central
atoms of the system —

260 RECENT ADVANCES IN ORGANIC CHEMISTRY
K . CH : CH . CH : CH . E E . CH : OH . CH : CH . R

Br— Br Br Br

E.CH.CHrCH.CH.E

I 1 I I
Br Br

The most striking application of the Thiele theory, however,
is found in the case of the benzene ring. If we write down the
Kekule formula for benzene, and fill in the partial valencies in
the usual way, we arrive at the following figure :—

An examination of this system will show that it forms
a closed series of conjugated double bonds. In other words, it
can be written as shown below, and no free partial valencies
exist in the system. Hence the impossibility of producing
addition products with benzene under ordinary conditions.

Though the theory of partial valencies has very widespread
application, it is not absolutely accurate, for several cases are
known in which it is not in accordance with the results of
experiment. We may mention one or two of these, without
laying too much stress upon them.

Harries * has shown that unsaturated aldoximes or ketoximes
may be reduced to unsaturated amines. Thiele himself mentions
a case observed by Bredt and Kallen 2 in which hydrocyanic

1 Harries, Annalen, 1903, 330, 185.
9 Bredt and Kallen, Annalen, 293, 338.

SOME THEORIES OF ADDITION REACTIONS 261

acid adds on to cinnamylidene-malonic acid by simple addition
to the double bond next the carboxyl radical. Hinrichsen and
Lohse observed that when cinnamenyl-cyanacrylic ester (I.)
is allowed to react with bromine it yields a bromide of the
formula (II.) shown below —

(I.) C6H5.CH:CH.CH:C

^ OOEt

(II.) C6H5 . CHBr . CHBr . CH : C

COOEt

Several other similar instances are known, and it appears
that, though in the main the partial valency theory is most
useful, in some cases it is necessary to take into account the
influence exerted upon the addition reaction by substituents
near the double bond.

This part of the subject has been investigated by Bauer,1
who was able to substantiate Nef s view 2 that addition reactions
may be influenced in this way. Bauer showed that if we
accumulate phenyl, carboxyl, or carbethoxyl groups or bromine
atoms in the neighbourhood of a double bond, bromine is not
easily taken up by the double linking. For instance, if we take
the general formula —

B R

and make KI a carboxyl group, bromine will be added on,
unless the other B groups are bromine atoms, or bromine atoms
with some methyl groups. Methyl groups alone do not hinder
the addition. Thus we get addition of bromine in the case
of acrylic acid, a- and /3-bromacrylic acids, crotonic and isocro-
tonic acids, dimethylacrylic acid, tiglic acid and trimethylacrylic

1 Bauer, Ber.t 1904, 37, 3488.

2 Nef, Annalen, 1898, 298, 208.

262 RECENT ADVANCES IN ORGANIC CHEMISTRY

acid. No addition of bromine takes place in the cases of tri-
bromacrylic and dibromcro tonic acids.

If we replace EI and E4 by carboxyl groups, addition of
bromine remains possible as long as E2 and E3 are not either
bromine atoms or methyl groups. Thus bromine will attack
maleic, fumaric, mesaconic, or bromomaleic acid, but it will
not attack the double bonds of dimethylf umaric, dibromofumaric,
or bromo-mesaconic acid.

If EI is a phenyl radical, and the other three Es methyl
groups, the compound takes up a molecule of bromine. If KI
and E4 are phenyl groups, and one of the remaining Es is a
hydrogen atom, the compound will react with bromine ; but if
in addition to the two phenyl groups we introduce two bromine
atoms, the additive power ceases. Thus while addition takes
place in the case of stilbene, methyl-stilbene, or bromo-stilbene,
it fails in the case of dibromostilbene.

If for EI and E2 we substitute phenyl groups, while the
other two Es are hydrogen atoms or alkyl groups, we enter a
new phase; for now we have first an addition reaction, and
then a spontaneous loss of hydrobromic acid, leaving us with a
bromo-substituted unsaturated compound. In this way behave
diphenyl-ethylene, diphenyl-propylene, diphenyl-methyl-pro-
pylene ; but no addition of bromine takes place in the cases of
diphenyl-bromo-ethylene, diphenyl-bromo-propylene, or tetra-
phenyl-ethylene.

In a later paper l Bauer showed that, when placed near a
double bond, the phenyl group had a certain effect upon the
addition of bromine, the carbethoxyl group had more, while
the nitrile group had the strongest influence of the three.

It was also found that the influence of the phenyl group
was weakened by nitration, a nitro-group in the meta-position
having least effect, and one in the ortho-position the most
influence.

Bauer showed further that the addition of bromine to the
double bond is a reversible reaction, equilibrium being attained
at different stages according to the effect of the substituents
introduced into the molecule.

It is a curious fact that substituents which influence the

1 Bauer, /. pr. Chem., 1905, II. 72, 201 ; Bauer and Moser, Ber.t 1907,
40, 918.

SOME THEORIES OF ADDITION REACTIONS 263

addition of bromine have a parallel effect upon the dissociation
constant of acids, the effect of a phenyl group in the one
case, for instance, being less than that of a cyanide radicle,
and the same holding good in the case of the dissociation
constants.

Somewhat similar results have been obtained by Klages 1
in the course of his researches on the reduction of styrolene
derivatives.

It will be noticed that in all the foregoing views the
question is treated purely from a static standpoint ; the double
bond is regarded more or less as a kind of hook which can
fix itself upon any atom wandering in the neighbourhood.
Stewart's view differs from the others in that it concerns itself
more with the dynamics of intramolecular change than with
the purely static side.

In the first place, we may give a brief account of the
chemical evidence upon which this view is based. Stewart 2
first showed that the reactive power of the carbonyl group in
acetoacetic ester greatly exceeded that of the carbonyl radicle
in acetone. Later3 he proved that the carbonyl radicle of
acetone dicarboxylic ester was even more reactive than that of
acetoacetic ester. On the other hand, the carbonyl groups in
Isevulinic ester and in acetonylacetone were much less reactive
than those of acetone. Acetylacetone proved to be more re-
active than either. Pinacoline was the least reactive of all
the ketones examined.

Taking acetone as the highest substance in the "slight
reactivity" class we can arrange the two sets of compounds
thus —

Low Reactivity. High Reactivity.

4 I Acetone ^ i (Acetone)

| Lsevulinic ester § Acetoacetic ester

g Acetonylacetone b Acetylacetone

p v Pinacoline « v Acetone dicarboxylic ester

And when we examine the " reactive " group we find that all
the substances which it contains are tautomeric bodies, which
are capable of yielding sodium derivatives, and whose methylene
hydrogen atoms are easily replaced by halogens. In other

1 Klages, Ber., 1903, 36, 3584; 1904, 37, 1721, 2301.

2 Stewart, Trans. Chem. Soc., 1905, 87, 185.

3 Stewart, Proc. Chem. Soc., 1905, 21, 78.

264 RECENT ADVANCES IN ORGANIC CHEMISTRY

words, not only is the carbonyl group in each of them very
reactive, but the hydrogen atoms of the neighbouring methylene
group share this activity. On the other hand, none of the
" weakly reactive " group are strongly tautomeric, and their
hydrogen atoms are not acidic to any great extent.

Now, in the case of the substance acetoacetic ester, for
instance, it is obvious that the hydrogen atoms in the methylene
group must be very closely related to the oxygen atom of the
carbonyl group ; for if they were not, it is difficult to understand
how the sodium derivative of such substances is derived from
the hydroxylic form of the compound. Without actually sup-
posing that the hydrogen atom wanders freely between the
carbon and oxygen atoms, as in the Laar hypothesis, it must be
admitted that the oxygen atom exercises some influence upon
the hydrogen atone, and vice versa. This influence will be most
strongly marked in the case of acetone dicarboxylic ester, but it
will be perceptible even in simple ketones.

Let us next consider the effect which this influence will
have upon the various atoms in the molecule. If we take the
case of the grouping —

(1) (2)
R— C— CH— X

O H

(3)

it is obvious that the affinity of the carbon atom (1) is occupied
in part by the group K, part by the carbon atom (2), and the
remainder is devoted to saturating the affinity of the oxygen
atom. Similarly, the affinity of the carbon atom (2) is distri-
buted between the group X, the carbon atom (1), and the
hydrogen atoms. But since there is some relation between the
oxygen atom and the hydrogen atom (3), some of the affinity
of these two atoms must be used up in mutual attraction, and
the more tautomeric the compound is, the more affinity will
thus be employed.*

Eegarded in this way, the molecule would represent a closed
system, the affinities of whose atoms are mutually saturated.

* Tautomeric is perhaps not the correct word; what is meant is that the
affinity- exchange between the two atoms will be greatest in the case where the
methylene hydrogen atoms are most strongly acidic.

SOME THEORIES OF ADDITION REACTIONS 265

But if we take into account the intramolecular motions of atoms,
the case at once assumes a different aspect. It is obvious that
the influence exerted by the hydrogen atom (3) upon the
oxygen atom will not be constant, but will vary according to
the distance between these atoms. If we assume, as is usually
done, that the atoms within a molecule vibrate in closed paths
about relatively fixed centres, it is evident that the hydrogen
atom will be now approaching, now retreating from the oxygen
atom. Every approach will ent:til a rearrangement of the
affinity of the two atoms, and another rearrangement of affinity
will take place during their retreat from each other. Stewart
considers that this rearrangement of affinity of these two atoms
is the cause of their chemical activity. In a closed or stable
system of forces the introduction of a new element is difficult ;
but if a system, be in a state of unstable or continually varying
equilibrium it is more readily amenable to change.

Now, in tautomeric compounds, since the oxygen and
hydrogen atoms exert great influence upon each other, this
redistribution of affinity will be much more strongly marked
than in the case of substances like acetone, whose hydrogen
atoms are only very weakly acidic ; and when we replace the
hydrogen atoms of acetone by methyl groups we shall still
further lessen the possible influence upon the oxygen atom,
which serves to make clear the very low reactivity of the
carbonyl group in pinacoline. Support is lent to Stewart's
views by the evidence of the absorption spectra of the
above-mentioned ketonic compounds which were examined by
Stewart and Baly.1

We must now turn to the question of the a-diketones and
quinones, which also contain very reactive carbonyl groups.
When the reactivity of the carbonyl group in pyruvic ester
was determined by Stewart's method,2 it was found to be
more active than any of those previously dealt with. Now,
in this case, we can hardly suppose that there is any great
attraction between the hydrogen atoms of the methyl radicle
and the oxygen of the carbonyl group : the chemical behaviour
of the substance gives us no right to draw any such

1 Stewart and Baly, Trans. Chem. Soc., 1906, 89, 489 ; of. Stewart and Baly,
Trans. Chem. Soc., 1906, 89, 618.

2 Stewart, Tran*. Chem. Soc., 1905, 87, 185.

266 RECENT ADVANCES JN ORGANIC CHEMISTRY

conclusion ; and the spectroscopic evidence, as far as it can be
considered relevant in a purely chemical question, tends to
disprove the existence of any such mutual influence. We are
therefore forced to a new point of view.

. If we assume that instead of a mutual attraction between
the oxygen and hydrogen atoms we have a similar attraction
between the oxygen atoms of the two carbonyl groups in
pyruvic ester, we should be able to explain how in that sub-
stance a redistribution of affinity is going on which to some
extent will resemble that in acetoacetic ester. But we may
go even further in this case, and assume that as an extreme
form of the intramolecular vibration we have almost a rearrange-
ment of bonds such as is expressed in the following formula : —

CH3— C— C— OEt CH3— C=C— OEt

II II II

0 0 0—0

In pyruvic ester it is doubtful whether this change ever
takes place ; we have only spectroscopic evidence in support of
it, not chemical. But there is a parallel case in which we can
bring actual chemical evidence in support of this attraction
between the two oxygen atoms. Wills tatter and Miiller,1 by
oxidizing catechol with silver oxide, have succeeded in isolating
ortho-benzoquinone in two isomeric forms, to one of which they
ascribe the dicarbonyl formula (I.), while the other they suppose
to have the structure (II.) —

CH CH

^ \ / \

HC C=0 HC C— 0

II I II I

HC 0=0 HC C— O

v v

It is well known that the para-positions of the benzene
ring are closely related to each other, more closely than
the ordinary structural formulae indicate; and in the case of
the quinones, this connection is very strongly marked in the

1 Willstatter and Muller, Ber., 1908, 41, 2580,

SOME THEORIES OF ADDITION REACTIONS 267

reactions of the two carbonyl radicles. Now, quinone itself is
a tautomeric body which reacts as if it had either of the two
structures shown below —

C C

/\ /\\

HC CH HC 0 CH

II II II |

HC CH HC O

\/ \J>

C C

(I.) (II.)

Hence it is obvious that in this case we actually have a change
taking place which converts the compound (I.) into the com-
pound (II.), and vice versa. This change is perfectly analogous
to that which we have already written down in the case of
pyruvic ester, though in that compound we had no chemical
evidence for the two formulse such as we have in the case of
quinone. The conversion of the quinone (I.) into the quinone
(II.) and its reverse would produce that redistribution of
affinity upon the oxygen atoms which we have postulated as
the cause of the chemical activity in carbonyl groups ; and the
vibrations of the benzene ring itself suffice to explain why this
conversion takes place, for in one phase we should expect to
find the carbonyl groups near to each other, while in another
phase they may be far apart, and therefore unable to exert
much influence upon each other.

Analogous ideas have been applied by Baly l to the case
of the colour of some nitrogen compounds, but as we are
dealing purely with the chemical side of the question here we
need not enter into any discussion of this physical property.

Forster 2 suggests that we might regard the oxygen atom
as changing its valency in these cases, so that the extreme

1 Baly, Edwards and Stewart, Trans. Chem. Soc., 1906, 89, 514; Baly,
Marsden and Stewart, ibid., 966.

2 Forster, private communication to the author.

268 RECENT ADVANCES IN ORGANIC CHEMISTRY

phase of the vibration in the case of camphor-quinone, for
example, might be written thus —

C=0

C8H14

and this view appears to have much to recommend it.

By means of the foregoing hypothesis, then, we are enabled
to give some explanation of a series of apparently quite dis-
connected facts: the reactivity of aliphatic ketones and diketones,
the tautomeric power of these substances, the sensitiveness of
the hydrogen atoms in the grouping — CO . CH2 . CO — , the
extreme chemical reactivity of the quinones ; and by a slight
extension of the hypothesis we could explain also the ease
with which the a-hydrogen atoms of acids are replaced by
halogens.

The foregoing brief account of the various theories of
addition reactions only serves to throw into relief the insuffi-
ciency of our present views on this subject. On the one
hand, we have ideas which are so vague as to convey very
little meaning, while on the other we have mechanical hypo-
theses which are too inelastic to cover anything but a very
narrow field. The most useful of all the suggestions hitherto
put forward, Thiele's partial valencies, deals rather with the
facts themselves than with any explanation of them, and takes
little account of subtle differences in reactivity. The field is
tempting to the theorist, however, and perhaps before long we
may have some view which will combine the advantages of all
the present hypotheses without their drawbacks.

CHAPTER XIII

UNSATURATION

WHEN we examine the matter closely, we find that the founda-
tions of theoretical organic chemistry are a series of labels by
means of which we endeavour to conceal our ignorance of the
fundamental phenomena of the subject. Of these labels,
none is used more indefinitely and at random than the word
" Unsaturation." It seems not without some interest, there-
fore, to examine the various phenomena which are usually
ascribed to the presence of this property, and to see how far
we can attain to some clear idea of what we mean by the
word.

In the first place, let us ask ourselves what we mean by
an unsaturated compound. The picture which is formed in
our mind by these words usually represents two molecules
uniting together, and one of these we are accustomed to call
an unsaturated substance. But before going further we are
faced by a difficulty, for there seems no reason why we should
consider one of the two molecules unsaturated and the other
saturated. For example, if a molecule of bromine unites
with a molecule of ethylene, we call ethylene an unsaturated
hydrocarbon, but we do not regard the bromine molecule as
unsaturated in the same sense. If we examine the matter
more closely, however, the difference between the two cases
becomes clear. When ethylene takes up an atom of bromine
the ethylene molecule is not completely disrupted ; part of it
remains as it was, for the two carbon atoms are still united,
and each bears the same number of hydrogen atoms as before.
With the bromine molecule, however, no trace of the original
structure remains. Evidently our idea of an unsaturated com-
pound must be extended ; it is no longer sufficient to say that
it is " a molecule capable of uniting with another molecule " ;
but we must add, "without a disruption of its original

270 RECENT ADVANCES IN ORGANIC CHEMISTRY

structure." This definition covers practically every case which
has any claims to be considered ; and we may therefore adopt
it and proceed to inquire if we can distinguish further between
the various classes of substances which come within the
definition.

The simplest type of an unsaturated compound with which
we can deal is a component of a double salt. Here the amount
of unsaturation is very slight, for we may decompose the
saturated body (double salt) into its components again by a
mere lowering of temperature.

The second class of unsaturated compounds includes those
in which the addition of new atoms takes place at one atom
only, as, for example, trimethylamine, dimethyl-sulphide,
dimethyl-pyrone, etc. In this case the least possible change
in the general structure of the molecule takes place during the
addition reaction.

The third class of unsaturated compounds contains those
bodies which are capable of uniting with two atoms, but in
which addition takes place at two adjacent atoms. The
ethylene series, the ketones, and the nitriles are instances of
this type.

There is another class of bodies which, while resembling the
last-mentioned one, in so far as their capability of adding on
only one pair of atoms is concerned, differs from it in the
manner of addition ; for, instead of the two new atoms attaching
themselves to two adjacent atoms, as in the ethylene class, in
this new series they attach themselves to non-adjacent carbon
atoms. The polymethylenes are a case in point.

Finally, we come to the acetylene class, in which we are
able to unite four new atoms to two of the carbon atoms of the
unsaturated compound.

Thus we have divided unsaturated bodies into the following
five classes : —

1. Components of molecular compounds.

2. Compounds of mono-valent iodine, divalent sulphur,

selenium, tellurium, oxygen, etc., trivalent nitrogen,
phosphorus, antimony, etc.

3. Compounds containing groups like C : C , C : N , C : 0.

4. Cyclic compounds.

5. The acetylenes.

UNSA TURA TION 27 1

Of course, it is quite easy to multiply the possibilities by
combining in one molecule representatives of each class, as in
the case of mesityl oxide, for instance ; but if we reduce the
question to its simplest form, the above series will serve as a
mode of classification.

Before entering into a consideration of these classes, how-
ever, we must deal with two other points which arise. What
we call an unsaturated substance may be unsaturated with
regard to one agent, and quite saturated towards another.1 For
instance, if we take the substances in Class 3, though all of
them are unsaturated with respect to nascent hydrogen, they
differ in their behaviour towards bromine, ammonia, or water.
Again, it is sometimes found that a compound may behave as
a saturated or an unsaturated substance according to the con-
ditions under which reactions are carried out. For example, in
sunlight benzene forms addition products much more easily
than in the dark. Thus there are fine differences for which we
have no corresponding technical terms.

It would occupy too much space were we to enter into any
detailed examination of the differences in physical properties
between saturated substances and the unsaturated bodies from
which they have been prepared. There is hardly a single
physical property which remains common to the two groups.
Melting-point, boiling-point, refractive index, optical rotatory
power, absorption spectrum, magnetic rotation, crystalline form,
electrical conductivity, and a host of other properties are all
changed by the addition of as many atoms as the unsaturation
requires.

The chemical effects of unsaturation are hardly less marked.
Leaving out of consideration the chemical difference implied in
the fact that the unsaturated compound is capable of adding on
more atoms, while the saturated one is not, there are many
other differences which the presence or absence of unsaturation
in. the molecule brings into view. For example, if we take a
saturated aliphatic acid and the corresponding unsaturated
substance in which the double bond lies next the carboxyl
group, the saturated acid will esterify with much greater ease
than the unsaturated one.2 Again, unsaturation may call into

1 See Vorlander, Annalen, 1902, 320, 66.

2 Sudborough and Koberts, Trans. Chem. Soc., 1905, 87, 1840 ; Sudborough
and ThomaB, ibid.t 1907, 91, 1033.

272 RECENT ADVANCES IN ORGANIC CHEMISTRY

being a peculiar type of isomerism, of which the best example
is found in the case of maleic and fumaric acids. Further, in
the case of the acetylene series, the presence of unsaturation so
alters the chemical characters of the hydrogen atoms that they
become replaceable by metallic atoms ; while, if we accumulate
acetylenic linkages in a compound, it may become so unstable
as to decompose with explosive violence. These few examples
will suffice to illustrate the very varied influences exerted on
the properties of compounds by unsaturation.

We may now turn to the question of the relative stability
of various unsaturated compounds. In the first place, it is
obvious that the amount of energy which is used up in satu-
rating the component of a double salt cannot be very great ; for
if it were so, the double salt would not be decomposed again
into its components with the ease which is found in practice.
In the second class, we obtain saturated compounds by calling
into existence some latent affinities on the sulphur or nitro-
gen atoms. Now, these new salts — sulphoniurn, ammonium,
phosphonium, or whatever they be — can, in many cases, be
broken down into the unsaturated substance again by very
simple means. For instance, merely by heating the quaternary
ammonium salts we can obtain the amido-compounds from
which we started. Thus, though we have here a set of sub-
stances more stable than the double salts, still the increase in
stability is not very great. When we come to the groups 3, 4,
and 5, the change from the saturated to the unsaturated body
can only be brought about by chemical means, so that in their
case we have passed into a new stage of the question.

There is another way in which we can look at the matter,
and, for the sake of simplicity, we may confine our investiga-
tion in the rest of this chapter to the cases of the carbon
compounds. If we take an ethylene derivative and compare it
with the isomeric polymethylene, we find that the former is
much more readily attacked by reagents than the latter; in
other words, the ethylene type is more unsaturated than the
polymethylene. Thus, while ethylene compounds are almost
instantaneously oxidized by permanganate, the polymethy-
lenes are not so rapidly destroyed. The acetylene series is
even more sensitive to oxidizing agents than the ethylenes.

A somewhat interesting point arises when we combine in

UNSA TURA TION 273

one molecule two different types of unsaturation, and then
endeavour to find out which of them is the more readily satu-
rated. For example, if we take the case of mesityl oxide, we
have in one molecule the double bond between two carbon
atoms, and the other double bond between a carbon and an
oxygen atom —

(CH3)2C
II
CH

CH3— C

II
0

These two double bonds are of different types, and hence we
should expect to find some differences between their chemical
activities. In the first place, of course, we find that the one
bond will react with halogen acids, which do not attack the
carbonyl group. But if we leave out of account such differ-
ences and confine ourselves to the action of those reagents
which are capable of reacting with both linkages, the results
are sufficiently striking. If we reduce mesityl oxide by means
of weak alkaline reagents, such as sodium amalgam or alumi-
nium amalgam, the carbonyl group remains intact, while the
double bond is opened up. Two molecules of the ketone unite
together to form a saturated diketone —

(CH3)2C.CH2.CO.CH3
(CH3)2C.CH2.CO.CH3 •

(In the case of aliphatic ketones this diketone further condenses
to a cyclic compound, in this instance desoxy-mesityl oxide _

(CH3)2C -- CH2\

(CH3)2C- -C-CO . CH3

while in the aromatic series the reaction may be stopped at the
first stage.)

When we use, as a reducing agent, sodium in aqueous ether,
the reaction takes quite a different course, for here both the

274 RECENT ADVANCES IN ORGANIC CHEMISTRY

carbonyl and the ethylene linkages are attacked simultaneously,
giving the saturated alcohol —

(CH3)2CH . CH2 . CH(OH) . CH3

No method has yet been discovered by which we can reduce
the carbonyl group of mesityl oxide without destroying the
ethylene linkage as well.

But if we take the case of another ketone, such as —

CH2:CH.CH2.CO.CH3

we shall find no difficulty whatever in reducing it with sodium
in alcohol or aqueous ether to the unsaturated alcohol —

CH2 : CH . CH2 . CH(OH) . CH3

If we examine the structures of the two substances we find
that in the case of mesityl oxide we have a conjugated double
bond, while in that of the second ketone we have two unconju-
gated linkages ; thus in the case of mesityl oxide we cannot
attack one bond without tampering with the other, while in
the second case we can reduce either separately. If we examine
the progress of the reaction which we should expect from Thiele's
hypothesis in the case of mesityl oxide, we find that in the first
place, hydrogen adds on to the two end partial valencies, as
shown below —

(CH3)2C ...... H (CH3)2CH

II I

H.C, CH

I )

0 CH3 . \j

II I

0 ...... H OH

But this new body is merely the enolic form of the ketone —
(CH3)2CH.CH2.CO.CH3

into which it will rearrange itself at once. Thus it is easy to
understand why the carbonyl group is never attacked first
when it is conjugated with another double bond of a different
nature.

Again, ammonia is an agent which is capable of acting both
upon carbonyl groups and on ethylenic linkages, but if we allow
it to react with mesityl oxide it attacks only the double bond

UNSA TURA TION 27 5

between the carbon atoms and leaves intact the carbonyl
radicle —

(CH3)2C (CH3)2C.NH2

II + NH3 = |

CH3.CO.CH CH3.CO.CH2

Mesityl oxide. Diacetonamine.

The matter becomes a little clearer when we consider the
action of hydroxylamine upon mesityl oxide.1 If the action is
allowed to take place in a methyl alcoholic solution in presence
of sodium methylate, the chief product is the substance formed
by the addition of hydroxylamine to the double bond —

(CH3)2C . CH2 . CO . CH3
NH.OH

But if, on the other hand, we take hydroxylamine hydrochloride
and after exactly neutralizing it with sodium carbonate allow it
to act upon an alcoholic solution of mesityl oxide, we get the
usual carbonyl group reaction, and mesityl oxime is formed —

(CH3)2C : CH . C(NOH) . CH3

Thus in alkaline, solution the ethylenic bond is stimulated into
activity, while in neutral solution the carbonyl radicle appears
the more reactive of the two.

The influence of the conjugated double bond makes itself
felt also in the cases of propenyl and allyl methyl ketones.
Blaise 2 has examined these two isomeric substances, and finds
that when treated with one molecule of hydroxylamine in a
neutral solution they are both converted into oximes —

CH3 . CH : CH . CO . CH3 -> CH3 . CH : CH . C(NOH) . CH3
CH2 : CH . CH2 , CO . CH3 -> CH2 : CH . CH2 . C(JSTOH) . CH3

But if we treat them with two molecules of hydroxylamine the
results are different ; allyl methyl ketone reacts as in the last
case, giving the oxime shown above, while propenyl methyl

1 Harries and Lellmann, Ser.t 1897, 30, 230, 2726; Harries and Jablonski,
ibid., 1898, 31, 1371; Harries, Annalen, 1904, 330, 191.

2 Blaise, Bull soc. chim., 1905, III. 33, 42.

276 RECENT ADVANCES IN ORGANIC CHEMISTRY

ketone reacts with two molecules of hydroxylamine to give the
hydroxylamineoxime derivative shown below —

I
CH.NH.OH

CH3.C: NOH

Thus the conjugation of the ethylene and carbonyl bonds
increases the activity of the ethylenic linkage in this case also.
Blaise showed that exactly similar results were obtained with
semicarbazide, the allyl ketone forming a semicarbazone, while
the propenyl ketone, in virtue of its conjugated bonds, took up
a second molecule of semicarbazide to form a semicarbazide-
semicarbazone.

Posner 1 has studied the matter very fully in order to find
what effect various groups exert when placed near the
ethylenic double bond. In the first place, he proved that the
ethylenic linkage alone was capable of taking up hydroxyl-
amine and mercaptans, so that this addition capacity does not
depend entirely upon the proximity to the carbonyl group.
When unsaturated acids were used, he found that the activity
of the double bond was weakened if the carboxyl group was
placed in its vicinity. Thus neither maleic nor fumaric acid
can be induced to combine with hydroxylamine except to
form the usual salts; phenyl-isocrotonic acid, on the other
hand, in which the ethylene linkage is not conjugated with the
double bond of a carboxyl group, takes up a hydroxylamine
molecule with special ease, a/3 -unsaturated monocarboxylic
acids give with hydroxylamine a-oximino-acids, whilst a/3-un-
saturated ketones form /3-hydroxylamine derivatives.

Thus we cannot say definitely that the ethylenic linkage is
more or less active than the carbonyl bond ; for the matter is
influenced in different ways by the reagent employed, the
solvent used and the relative position of the two double bonds
in the molecule. In other words, " unsaturation " is not a
definite, measureable thing which we can predict in any case
from the behaviour of the " unsaturated " substance in other

1 Posner, Ber., 1901, 34, 1395; 1902, 35, 799; 1903, 36, 4305; 1905, 38, 646;
190G, 39, 3515 ; 1907, 40, 218 ; Posner and Oppermann, ibid., 1906, 39, 3705.

UNSA TURATION 277

circumstances ; it is rather something kinetic, something which
is extremely sensitive to external forces, and which in its turn
can play a part in influencing the chemical action of groups
which it does not apparently affect directly.

As an example of this latter property we may quote the case
of the Vorlander Eule.1 Vorlander has pointed out that we can
consider both acids and alcohols as derived from water by
substitution. In the case of acetic acid we substitute an acetyl
group for one of the hydrogen atoms of water, while ethyl
alcohol is formed from water by the substitution of an ethyl
group for a hydrogen atom.

H CH3.CH2 CH3.CO

O 0 O

/ ^ / /

When we examine the chemical behaviour of the hydrogen
atom in each case, we find that in the acids it has a much greater
activity than in the alcohols. The origin of this difference
obviously lies in the difference between the acyl and alkyl
groups to which the hydroxyl radicle is united. The question
is commonly dealt with by labelling the acyl group " electro-
negative," and treating the label as an explanation. But, as
Vorlander pointed out, this case is only one example of a general
rule. If we represent non-metallic elements by E, and write
down the following series : —

1234

H.E.E:E

123

H.E:E

12345

H.E.E.E-.E

we shall find that the hydrogen atom in the first line has a
greater reactivity than those in the second and third lines ; in
the first case the double bond between two E atoms lies in the
3 : 4 position to the labile hydrogen atom, while, where the
double bonds are in the 2:3 or 4 : 5 positions the hydrogen
atom is not specially active. For example, the labile hydrogen
atoms in oximes, acids, phenols, diazo-compounds, and sulphuric
1 Vorlander, Ber., 1901, 34, 1633.

278 RECENT ADVANCES IN ORGANIC CHEMISTRY

acids are all situated as in the first type with respect to the
double bond —

4321 4321

E.CHrK.O.H OcC.O.H

I
B

4322 4321

R— N:N.O.H 0:S.O.H

CH-CH

C— 0— H

CH— CH

In acetoacetic ester and nitre-methane the hydrogen atoms
are doubly influenced —

0(4) 4 0

II II 2 1

CH3 . 0(3) 3 N— 0— H

I II

H— C H 4 CH2

I (2) (1)
EtO— 0(3)

0(4)

Further, when an acid or a ketone is brominated, the
halogen atom enters the nucleus in the position required by
this rule, i.e. it replaces the hydrogen atom in the o -position to
the carbonyl group —

1234 1234

Br . CH . C : 0 Br . CH . C : 0

E OH R R

There seems to be another influence at work in the case
of acidic hydrogen atoms; and as the matter appears to have
escaped notice hitherto, it may be well to call attention to it
in this connection.1 An examination of the formulae of most
substances which are capable of yielding metallic derivatives

1 Smiles, Trans. Chem. Soc., 1900, 77, 160.

UNSA TURA TION • ; 279

will show that the atom to which the labile hydrogen is
attached is capable of exerting a valency higher than that
which it exhibits in the acidic compound. For example, in
the following substances the oxygen and sulphur atoms are
divalent, while both oxygen and sulphur are capable of acting
as quadrivalent elements; carbon in acetylene acts as a
divalent atom, though its maximum valency is four; the
nitrogen atoms shown below are trivalent, but nitrogen can
act as a pentad ; iodine can act either as a mono- or a trivalent
element. The formulae are written with lines to show the
extra valencies.

H-I/

H-C— C—

H H— O— CH3 H— S— CH3

CH=CH
H— NH— CO . CH3 H— N

It will be seen that this is of more general application than
the Vorfander Kule, for it holds in the case of substances such
as ethylates, whose formation takes place though there is no
double bond in the molecule such as is required by Vorlander's
view.

We must now turn to another point of view. Hitherto we
have regarded unsaturation from the standpoint of addition
reactions, but we may now extend this a little. Suppose that
we have two isomeric substances, each capable of taking up
four bromine atoms, are these two bodies equally saturated or
are they not ? The question of unsaturation thus resolves itself
into one of stability. We cannot distinguish between the bodies
by the amount of bromine they take up, so we seek some other
criterion. Now, in the case of two substances, one of which
has a pair of conjugated double bonds, while in the other the
bonds are not so related, the second substance takes up the four
bromine atoms at once, but the first one takes them up two
by two. The action is thus more precipitate in the second
instance, and we should be tempted to consider the first substance
as the less unsaturated of the two. In fact, as Thiele put it,
the conjugated double bonds partially saturate one another.

28o RECENT ADVANCES IN ORGANIC CHEMISTRY

Further, when an unsaturated acid is brought into conditions
which allow it to undergo isomeric change, it is almost always
converted into the form which contains the ethylenic bond con-
jugated with that of the carboxyl group. Evidently, then, this
grouping must be the most exothermic, and therefore the most
saturated.

We may now sum up, as far as possible, the various points
which we have treated in the foregoing pages. We have shown,
in the first place, that unsaturation is not an intrinsic property
of any molecule. It depends largely upon the nature of the
outside reagent ; in order to have unsaturation we must have
two substances, each specially fitted to interact with the other.
In fact, the addition reactions of organic chemistry appear to
be an extreme case of the ordinary reactions of salt formation,
such as takes place in the case of ammonia and acid. Secondly,
the influence of the other (non-reacting) parts of the molecule
may play a very considerable part in any addition reaction, so
that we cannot ascribe the same meaning to every double bond
that we write down. For example, the ethylenic bond in
maleic acid must be chemically quite different from that in
mesityl oxide. Thirdly, just as unsaturation can be influenced
by neighbouring un saturations, it can in turn exert an influence
upon groups of atoms in its vicinity. And, finally, if we have
a series of unsaturations in a molecule they can be made to
rearrange themselves to form a more stable system.

It has thus been shown that the term "unsaturation"
covers a very wide and ill-defined field. Our knowledge of the
whole problem is very scanty at best, and we are handicapped
(and likely to remain so) owing to the fact that no one has the
faintest idea of what really lies at the back of the various
phenomena which we catalogue under this name.

CHAPTER XIV

CONCLUSION *

IT often happens that we meet with a series of apparently
related facts, and we are anxious to put forward some plausible
explanation which will make the connection between them
clear. Under these conditions we may proceed on either of
two alternative lines. For instance, on the one hand we may
put forward some general idea which, without troubling about
details, will allow us to regard the matter from a broad point of
view ; or, on the other, we may set up some mechanical model
which will, as far as possible, reproduce the phenomena we set
out to explain.

At the first glance, the former method seems the more
likely to lead near the truth ; but consideration will show us
that this is not the case. Suppose that our general idea covers
all the facts known at a given time, and is quite comprehen-
sible when considered in relation to these facts. Then let us
imagine that some new facts are discovered which do not quite
agree with the general idea. As a result, the general idea is
widened to include these facts, and thus it becomes more vague
than it was before. After this process has continued for a time,
the general idea is widened insensibly, until it ceases to have
any definite meaning. It eventually becomes a mere rag-bag
of views or an amorphous mass which can be squeezed to fit
any vessel. Naturally, also, it has ceased to have any value
from the scientific point of view — it can no longer stimulate
us in research, nor can it aid us in our classification of facts.

With a mechanical hypothesis, on the other hand, we have
something definite, which either does or does not fit the facts.
If it fits them, well and good ; but as soon as it ceases to agree

* This chapter was written at the suggestion of Professor Collie, who is
partly responsible for some of the ideas expressed in it.

282 RECENT ADVANCES IN ORGANIC CHEMISTRY

with our data we discover the inadequacy, aud can discard the
mechanical model, replacing it by another which is more in
accord with our increased knowledge. We cannot, as in the
case of the general idea, expand it and make it more vague, but
we can expand it while retaining its definiteness.

The difference between the two methods can be seen by
comparing the system of " energetics " put forward by Ostwald
and others with the modern structural (mechanical) theories
of organic chemistry. It is quite certain that no application
of the purely " energetic " view to organic chemistry could
ever have carried the science to the point at which it now
stands.

In the first chapter of this volume the question of the real
meaning of our modern structural formulae was touched upon,
and it was pointed out that Kekule and Couper differed with
regard to this matter. Up to the present time, there is no
doubt that the Kekule view has prevailed to a great extent.
Our formulae for organic compounds are reaction-formulas ; they
represent merely the behaviour of the substance when treated
with various reagents, and it is simply on this account that
our ordinary structural formulae are now failing to meet the
demands which are made upon them by recent work. If we
take the case of quinone as an example, we find that its formula
is written in either of two ways —

c c o

/ \ / \

HC CH EC CH

II II II |

HC CH HC CH

\ / \ /-

C C

each of which is a representation of its method of reacting with
a certain reagent. But neither of these formulae allows us to
foresee the fact that quinone monoxime will react as if it were
nitroso-phenol —

CONCLUSION 283

NOH N:0

11 I

C C

X\ S\

HC CH HC CH

II II I II

HC CH HC CH

v • v

II I

0 OH

The number of facts of this type which have accumulated in
recent years is very considerable, and the result of this increase
in knowledge has been somewhat remarkable. Instead of
attempting to bring their formulae into harmony with the facts,
organic chemists have been content to drag behind them a
lengthening chain of implications, which they read into a
formula; e.g. in the case of acetone and ethyl acetate we do
not distinguish in our formulae between the two carbonyl
groups, but we mentally interpret the two symbols differently.
Thus, at the present time, it is quite conceivable that a student
may be well acquainted with the meaning of all the ordinary
chemical symbols, but may be hopelessly at sea with regard
to the behaviour of a given compound, which to a more experi-
enced chemist is implicitly expressed in the formula which
misleads the student.

A concrete example will serve to bring out the amount of
unexpressed material which we read into the ordinary formula.
Let us consider the reactions of the unsaturated monobasic
acids in presence of dilute sulphuric acid. In the first place,
we assume that an addition of water to the double bond
occurs —

(CH3)2C HO (CH3)2C— OH HO

HC— CH2— CO CH2— CH2— CO

Now, we know from general experience that when one
hydroxyl group lies in the 1, 6-position to another in the same
chain, water is usually eliminated with ease; so we should
deduce that the next step in the process would be such an
abstraction of a water-molecule —

284 RECENT ADVANCES IN ORGANIC CHEMISTRY

(CH3)2C - OH HO (CH3)2C -- 0

I I -- > I !

CH2-CH2— CO CH2 - OH2— CO

The formation of this compound is actually what does take
place, so that in this case our implications are justified; but
let us apply the same series of ideas to another instance. Take
the case of vinyl-acetic acid (I.), which contains the double
bond in exactly the same position as in the other substance.
Applying our experience as before, we should deduce that the
final product on heating with dilute sulphuric acid would be
the lactone (II.). In practice no such substance is formed, the
product being the new unsaturated acid (III.).

CH2 ---- 0
CH2:CH.CH2.COOH | I CH3 . CH : CH . COOH

CH2— CH2— CO
(I.) (II.) (HI.)

But this does not bring us to the end of the possible reac-
tions of this class of substances, for if we take the case in
which two methyl groups are attached to another carbon atom
we find that the reaction follows yet another course —

CH2:CH.C(CH3)2.COOH

CH3 . CHOH . C(CH3)2 . COOH

CH3 . CH : C(CH3)2 + C02 + H20

Thus, our formulae have ceased to be true reaction formulae,
and merely serve to mislead us if we attempt to draw any
general conclusions from them.

But this by no means ends the confusion to which our
modern formulae have given rise. It is evident that, treated as
mere reaction formulae, they do not fulfil our requirements ;
but there is another side to the question upon which they are
quite as unsatisfactory. Until, say, twenty years ago, the
relation between chemical constitution and physical properties
had not been very thoroughly investigated, and it was imma-
terial whether or not we could bring the two sides of the
subject into harmony with our structural formula?. Now,

CONCLUSION 285

however, that is all changed. The study of the physical
properties of chemical compounds has made vast strides, even
within the last two decades, and the mass of material now at
our disposal in this branch of the subject is continually increas-
ing. At the same time, no attempt has been made (except
in a few cases like Baeyer's carbonium l bond) to broaden the
basis of our structural formulae, so as to allow us to merge into
them both the physical and chemical behaviour of substances.
Instead, we have the usual train of implications, which are not
expressed in the formulae we write down, but are left to be
inserted mentally.

Within the last few years, also, several authors have pointed
out a connection between the reactivities of substances and
certain of their physical properties. Such a relation has been
shown by Bruhl and Schroder in the case of refractive index,2
by Kauffmann 3 in that of magnetic rotation, and by Stewart
and Baly 4 in absorption spectra. In these cases we are
dealing with some " reaction-property," so that the matter
certainly deserves consideration along with the ordinary chemical
reactions.

When all is said and done, however, it cannot for the moment
be considered desirable that we should get rid of our present
style of formulae ; they represent so much, and are undoubtedly
more convenient than any substitute at present conceivable.
What is required is that we should endeavour to bring them
into harmony with reactions on the one side, and with physical
evidence on the other. If this could be done we should see our
way much more clearly in the subject, and should not be
misled by as many false analogies as at present is the case.

To-day we are apparently standing on the verge of a new
view of things, which may conceivably carry us as far in
advance of present-day structural chemistry as the Couper and
Kekule formulae carried the chemistry of the Type Theory ; and
though it is impossible to say exactly what the new view will
be, it is not beyond our power to show the foundations upon
which it must rest, and the problems which it must solve if it
is to constitute a real advance upon our present position. We

1 Baeyer, Ber., 1905, 38, 569.

2 Bruhl and Schroder, Zeit. physical. Chem., 1904, 50, 1 ; 1905, 51, 18, 513.

3 Kauffmann, ./. pr. Chem., 1903, II. 67, 334.

4 Stewart and Baly, Trans. Chem. Sue,, 1906, 89, 489.

286 RECENT ADVANCES IN ORGANIC CHEMISTRY

shall endeavour to present a sketch of these in the remainder of
this chapter.

It will be remembered that the chief basis upon which our
modern views of chemical structure were laid was the recogni-
tion of Frankland's doctrine that atoms had a constant valency.
For many years this view sufficed for chemists, but in the more
advanced thought of the present day, doubts have arisen as to
its truth ; and it seems very probable that in a few years' time it
will cease to be regarded with quite the same definiteness as now.
In one of the foregoing chapters it was pointed out that
when we write a double bond between two atoms, we do not
always mean the same thing. Thus the double bonds in the
cases of diphenyl-ethylene, ethylene, and fulvene certainly do
not resemble one another chemically; in the first case the
double bond is not attacked by bromine, which is taken up
easily by the double bond of ethylene ; but while the fulvene
series are oxidized by air, ethylene substances are not. Thus
we have an increase in unsaturation (or reactivity as regards
bromine and oxygen) as we go from diphenyl-ethylene through
ethylene to the fulvenes ; while we symbolize all three unions
between the carbon atoms of the double bonds in exactly the
same way. It is perfectly evident that the amount of reactivity
is different in these three cases, and therefore the "valency-
force," which gives rise to reactions, must be different also.

But it is not only in the case of the double bond that we
can trace this alteration in value of valencies ; we can discover
it in the case of single bonds as well. It is well known that if
we take bromo-benzene, the bromine atom is held to the carbon
atom of the nucleus more firmly than is the case in aliphatic
bromine derivatives. But if we nitrate the benzene ring, the
bromine in the aromatic bromine derivative becomes as labile
as that in the aliphatic one. This increase in reactivity can
be due only to some change in the force which holds together
the carbon and bromine atoms ; in other words, the " valency-
force " uniting bromine to carbon is stronger in bromobenzene
than in nitro-bromobenzene.1 Flurscheim2 has carried out

1 When the above paragraph was written in 1908, 1 was under the impression
that this was common knowledge. Dr. Flurscheim desires me to mention, how-
ever, that he published a paper on the point in 1906 (Ber., 39, 2016).— A. W.S.

2 Flurscheim, /. pr. Chem., II. 1902, 66, 329 ; see also Werner, Ber., 1906,
39, 1278.

CONCLUSION 287

some experiments by means of which he shows that this
variation in the value of the single bond is quite a general
property.

It may be supposed by some that if we accept these ideas
we shall be taking a retrograde step, and plunging ourselves into
a web of inconsistencies ; but surely it is not so ! At the time
of Frankland, chemists had not acquired those ideas of chemical
structure which we now possess, and which we cannot abandon
without having something better to take their place; con-
sequently, it was necessary for the science to go through a stage
in which valency was regarded as a fixed, unalterable force ;
without this guiding principle the work of the last forty years
would have been impossible. But we have now reached a stage
when we can look back and enlarge our views without running
the risk of losing hold of what we have acquired. Instead of
regarding a " bond " as a fixed unit, we can afford to regard it
rather as the sum of an almost infinite number of small forces ;
so that we can subtract from or add to its strength within
limits without bringing it out of the category of a " bond " or
valency. For example, if the force employed in uniting two
atoms together by means of a single bond be termed " F3," then
the quantity F will be negligible in comparison with the force
of the single bond. But it is quite conceivable that this small
force F would be sufficient to cause a difference of reactivity
according as it were added to or subtracted from the force F2.
Thus the two forces expressed by —

F2 + F and F2 - F

would not differ appreciably in their capacity for uniting two
atoms, and certainly would not be so different as to allow the
first atom to unite with two others ; yet at the same time they
would be sufficiently different to produce a change in reactivity
of an atom attached to another by one or other of them.

There is another point of view which we must examine. If
we are going to broaden the basis upon which our chemistry
rests, we must be prepared to include the physical1 as well
as the chemical properties of substances in our classification :
and naturally for an explanation of the physical properties of

1 The relations between chemical constitution and physical properties have
been fully dealt with by Dr. Smiles in his book on this subject. Those interested
in the question should consult especially the concluding chapter of his volume.

288 RECENT ADVANCES IN ORGANIC CHEMISTRY

substances we must go to the physicists, and from them we can
borrow as much of their theory as seems likely to help us in
our own branch of science. Within the last few years, physics
has been to some extent rejuvenated by the conception of the
electron ; and it may not be amiss to give a brief account of the
modern electrical theory of matter in order to point out how
it might be applicable to our own subject.1

According to the modern view of matter, the atom consists
of a " positive sphere " (i.e. a sphere throughout which positive
electricity is uniformly distributed), within which lie a number
of electrons. Excluding radio-active bodies from our considera-
tion, it may be said that the electrons in an atom are of two
kinds. The first type, "fixed electrons," are those which
remain always attached to the positive sphere ; the other class,
" detachable electrons," are more or less free to wander from
atom to atom. Now, when a detachable electron leaves one
atom and makes its way towards another, it is supposed to
carry with it the end of a "tube of force," or "Faraday tube" ;
so that when it penetrates into the second atom, and remains
there, the two atoms become united by this tube of force.

It is evident that what the chemist calls valency, the
physicist would regard as an accumulation of tubes of force.
Thus the weakening or strengthening of a bond could be
regarded either as a reduction or increase in the number of
Faraday tubes joining two atoms together, or as a decrease in
the strength of the tube of force uniting the one atom to the
other.

Since Faraday tubes are supposed to be mutually repellent,
this hypothesis furnishes us with a new way of regarding
Baeyer's Strain Theory. It is obvious that in a double bond
between carbon atoms, the Faraday tubes will be more numerous
and closely packed than in the case of the single bond. They
will thus be more strongly repelled from the line joining the
centres of the two carbon atoms, and a state of strain will be
set up just as Baeyer postulated. The same holds good in the
case of the acetylene derivatives also.2

1 Nelson and Falk (School of Mines Quarterly, 1909, 30, 179) have developed
this line of thought in their paper on " The Electron Conception of Valency in
Organic Chemistry," and have shown how isomerisrns of different types may all
be expressed in terms of electronic motion.

2 Dr. Fliirscheim has called my attention to a paper of his (J. pr. Chem., II.

CONCLUSION 289

On the physical side, it has been shown that the properties of
matter are very largely due to the movements of the electrons
within the positive sphere ; while on the chemical side, as we
have seen, the reactivity and valency of atoms depends upon
the detachable electrons which they contain. It is obvious
that these two phenomena are not independent ; but that,
on the contrary, a very close relationship must exist between
them. Further, physicists have shown us that some atoms are
capable of shedding electrons more readily than others ; which
may aid us in our correlation of physical and chemical
properties.

Thus it becomes clear that any future advance on the
theoretical side of organic chemistry must take into account
the following branches of the subject : —

I. The arrangement of atoms in space.

II. The mode of linkage of the atoms within the molecule.

III. The influence which two non-adjacent atoms in a

chain can exercise upon each other.

IV. The difference between ionic and non-ionic reactions.
Y. Those physical properties of organic compounds which

are due to electronic motions.

1907, 76, 190) in which, starting from somewhat different premises, he arrives at
similar conclusions.

BIBLIOGRAPHY

WITHIN the last decade the literature of organic chemistry
has become increased and specialized to a much greater extent
than before, and it may be useful to give in this place a brief
account of some of the more recent works in various branches
of the subject.

Beilstein's " Handbuch der Organischen Chemie," which is
now complete in nine volumes, is too well known to need any
description ; and the same may be said for Kichter's " Lexikon
der Kohlenstoffverbindungen," to which three supplements
have been published. The former work was completed in
October, 1906 ; while the third supplement to the " Lexikon "
is complete up to 1904. New editions of both works are in
preparation.

As regards practical methods, Lassar-Cohn's "Arbeits-
methoden fur organisch-chemische Laboratorien " has now
reached its fourth edition (1906-7), which exceeds the third
edition (1903) by no fewer than 607 pages. The index of the
new edition is much superior to that in the older volume. A
new work on somewhat similar lines : " Die Methoden der
organischen Chemie ; ein Handbuch fur die Arbeiten im Labo-
ratorium " has been brought out by Weyl. The first volume
and part of the second have been published (1909-10).

The historical side 'of the subject is treated by Schorlemmer
in his " Kise and Development of Organic Chemistry," but this
only carries the matter up to the year 1894; later develop-
ments are to be found in Ladenburg's " Entwicklungsgeschichte
der Chemie" (1907), Meyer's " Geschichte der Chemie" (1905),
Armitage's "History of Chemistry" (1906), or Hilditch's
" Short History of Chemistry " (1910). The history of modern
theories in organic chemistry has been dealt with by Pattison
Muir in his "History of Chemical Theories and Laws," and
also by Hilditch in his " Short History." The Chemical
Society's " Memorial Lectures," as well as the obituary notices

BIBLIOGRAPHY 291

in the Berichte der deutschen chemischen Gesellschaft may be
consulted by those who are interested in the personal side of
chemical development.

The number of annual summaries of work published con-
tinues to increase. By the side of the " Jahresbericht der
Chernie " and Meyer's " Jahrbuch der Chemie," we have now
the Chemical Society's Annual Eeports ; while a " Jahrbuch
der organischen Chemie" has been begun by J. Schmidt
(vol. i., 1907).

Turning to the modern text-books on the subject of organic
chemistry, the most important is, of course, Meyer and Jacobson's
" Lehrbuch der organischen Chemie " (vol. i., 1893 ; vol. ii. part
1, 1902 ; part 2, 1903). It is unfortunate that this work is not
yet complete, the volume on the heterocyclic substances being
still lacking. A new edition of the first volume is in prepara-
tion, part of it (dealing with the aliphatic substances and their
mono-derivatives) having already appeared (1907). A tenth
edition of the Kichter-Anschutz "Organische Chemie" has
appeared (vol. i., 1903 ; vol. ii., 1905), and the first volume of
the eleventh edition was published in 1909. The German
edition of Koscoe and Schorlemmer was completed some years
ago. Among more recent works J. Schmidt's " Kurzes Lehr-
buch der organischen Chemie " deserves special mention.

In recent years several volumes of essays on various
branches of organic chemistry have appeared. Of these, Lach-
man's " Spirit of Organic Chemistry " is the most original.
It contains an historical account of the development of various
outstanding researches, such as those on rosaniline dyes, the
sugars, maleic and fumaric acids, Perkin's reaction and aceto-
acetic ester. A French collection, published under the title
" Ke'cents Progres de la Chimie," has reached its third volume
(vol. i., 1904 ; vol. ii., 1906 ; vol. iii., 1908), and contains much
that is of interest to the organic chemist. Cohen's " Organic
Chemistry for Advanced Students," and Keane's " Modern
Organic Chemistry " (1907) probably should also be included
in this category. In his "Neuere Anschauungen auf dem
Gebiete der organischen Chemie" (1908) Henrich has dealt
with the development of the more important modern theories.

We must now mention some monographs on special
subjects, Since 1896 a collection of pamphlets has been

292 RECENT ADVANCES IN ORGANIC CHEMISTRY

issued by Ahrens under the title " Sammlung chemischer und
chemisch-technischer Vortrage; and recently a somewhat
analogous series has appeared under the name " Die Wissen-
schaft." We need not deal with these in detail here, as we
shall have occasion later to mention the numbers which
specially concern us.

Under the head of stereochemistry we find that in the last
ten or fifteen years a perfect flood of text-books and pamphlets
has been issued, which testifies to some extent to the interest
taken in this branch of chemistry. The most complete work of
reference on the subject is Bischoff and Walden's " Handbuch
der Stereochemie " (1894), with its two supplementary volumes,
BischofFs " Materialen der Stereochemie " (1904). In these are
abstracts of practically every paper that has been published on
the subject up to 1902, the abstracts being grouped according
to the year of publication, and very well indexed. Werner's
"Lehrbuch der Stereochemie" (1904) and Stewart's "Stereo-
chemistry" (1907) are the most complete text-books on the
subject. Smaller works are Hantzsch's " Grundriss der Stereo-
chemie " (1904), van't HofF s " Lagerung der Atome in Eaume "
(1908), Mamlock's "Stereochemie" (1908), and Wedekind's
booklet, "Stereochemie," in the Sammlung Goschen (1906).
Landolt's " Optical Kotating Power of Organic Substances "
(1902) deals with one section of the subject in great detail.
Among the numerous pamphlets we need only mention the
following : Wedekind and Frohlich, " Zur Stereochemie des
fiinfwertigen Stickstoffs" (1907); Meyerhoffer, " Gleichgewichte
der Stereomeren " (1906) ; Jones, " Stereochemistry of Nitro-
gen" (British Association Keport, 1904); Ladenburg " Ueber
Eacemie " (Ahrens Sammlung, 1903) ; Schmidt, " Einfluss der
Kernsubstitution auf die Eeaktionsfahigkeit aromatischer Ver-
bindungen" (Ahrens Sammlung, 1902); Scholtz, "Einfluss der
Eaumerfiillung auf den Verlauf chemischer Eeaktionen " and
" Die optisch-aktiven Verbindungen des Schwefels, Selens,
Zinns, Siliziums und Stickstoffs" (Ahrens Sammlung, 1898
and 1907).

Three other papers in the Ahrens Sammlung deserve
mention. Marckwald, in one of them, gives a sketch of the
" Benzoltheorie " to 1897 ; while a second is devoted by Goose
to the "Beziehung der Benzolderivate zu den Verbindungen der

BIBLIOGRAPHY 293

Fettreihe " (1898). W. Wislicenus, in the same series, gave a
pamphlet, " Ueber Tautomerie " (1897) ; while the same subject
is treated by Lowry in the British Association "Keport on
Dynamic Isomerism" (1904).

The literature of the terpenes is an ever-increasing one.
Heusler's "Die Terpene" was published in 1896, and is now,
to a great extent, out of date, as is also Scholtz's pamphlet, "Die
Terpene," in the Ahrens Sammlung (1896). Gildemeister and
Hoffman's "Ethereal Oils" (1900) deals only with the occur-
rence and commercial side of the question, and does not trench
upon the problem of terpene constitutions. In Meldola's work
on " The Chemical Synthesis of Vital Products " (1904), the
occurrence and syntheses of many terpene derivatives are
described. A very complete account of all the alicyclic series
is given by Aschan in his work, " Die Chemie der alicyklischen
Verbindungen " (1905), while the terpenes themselves are very
fully described by Semmler in his work on " Die setherischen
Oele" (1906). In both of the two last works the authors have
given very clear and complete accounts of terpene structures.
The chemistry of camphor has been dealt with by Lapworth in
the 1900 British Association Eeport on " The Constitution of
Camphor," and by Aschan in "Die Konstitution des Kamphers"
(1903), which was afterwards incorporated in his "Chemie der
alicyklischen Verbindungen." Wallach has republished his
papers on the terpenes in a volume " Studien ueber die
Terpene" (1909); and a condensed description of this class
of compounds has been written by Barthelt (1909).

The chemistry of the alkaloids has not produced any ex-
tensive literature apart from the papers in the journals. The
older books, such as Guareschi's " Die Alkaloide," are out of
date, to a very great extent, as regards the constitutions of the
alkaloids, though they are still useful as far as descriptions
of physical and chemical properties are concerned. Scholtz's
pamphlet on " Die kiinstliche Auf bau der Alkaloide " (Ahrens
Sammlung, 1897) contains a good account of the state of the
subject at that period. Later, Pictet's " Vegetable Alkaloids "
(1904) brought the subject more up to date. We are most in-
debted, however, to J. Schmidt, who has brought out from time
to time small volumes upon the state of alkaloid chemistry.
Three of these have already been published : " Ueber die

294 RECENT ADVANCES IN ORGANIC CHEMISTRY

Erforschung der Konstitution wichtiger Pflanzenalkaloide "
(1900); "Die Alkaloidchemie in den Jahren 1900-1904";
"Die Alkaloidchemie in den Jahren 1904-1907." It is
probable that the series will be continued. The pharmaco-
logical side of the subject is dealt with by Frankel in his
" Arzneimittel Synthese" (1906).

With regard to the sugars, Tollens' "Kurzes Handbuch
der Kohlenhydrate " (1895) ; Lippmann's " Chemie der Zucker-
arten" (1895); or Maquenne's " Les sucres et principaux
derives" may be consulted. Fischer's papers have been re-
printed under the title " Untersuchungen Uber Kohlenhydrate
undFermente" (1909).

Fischer has republished his papers on the purines under the
title, "Untersuchungen in der Puringruppe" (1882-1906),
and those on the polypeptides in a volume called " Die Amino-
sauren, Polypeptide und Proteine" (1906). The following
works have appeared dealing with the problems of the proteins
and albumins : Mann, " Chemistry of the Proteids " (1906) ;
Schryver, "Chemistry of the Albumens" (1906); Schulz,
"Allgemeine Chemie der EiweissstofFe " (Ahrens Sammlung,
1906). A new series of monographs on biochemistry has
recently been issued under the editorship of Plimmer and
Hopkins. Up to the present, the following volumes have
been published : — Bayliss, " The Nature of Enzyme Action ; "
Plimmer, " The Chemical Constitution of the Proteins ; "
Schryver, " The General Character of the Proteins ; " Osborne,
"The Vegetable Proteins;" Armstrong, "The Simple Carbo-
hydrates and the Glucosides." Other volumes are announced
dealing with "The Development of Biological Chemistry"
(Hopkins) ; " The Carbohydrates and Polysaccharides " (Ling) ;
"The Fats" (Leathes); "Colloids" (Hardy); and "Alcoholic
Fermentation " (Harden).

The diazo-compounds have been described by Hantzsch in
his pamphlet " Die Diazoverbindungen " (Ahrens Sammlung,
1903) ; by Morgan in the British Association Eeport of 1902
on that subject ; and lastly, by Cain in his recent volume
" The Chemistry of the Diazo-compounds (1908).

Organic dyestuffs have been dealt with by Nietzki in his
" Chemie der organischen Farbstoffe " (1901), and later in
his pamphlet "Die Entwicklungsgeschichte der kiinstlichen

BIBLIOGRAPHY 295

organischen Farbstoffe" (Ahrens Sammlung, 1902). Cain and
Thorpe have recently brought out a work on " Synthetic Dye-
stuffs and Intermediate Products from which they are derived."
A small but comprehensive volume in the Sammlung Gb'schen,
Bucherer's "Die Teerfarbstoffe," gives a condensed account of
these substances. The constitutions and properties of the
naturally occurring dyes have been described by Kupe in his
"Chemie der natiirlichen Farbstofle" (Vol. I., 1900; Vol. II.,
1909).

The Ahrens Sammlung contains also some treatises on more
or less isolated questions of chemical interest, such as Hjelt,
" Ueber die Laktone " (1903) ; Wedekind, " Die Santoningruppe "
(1903) ; and Schmidt, " Ueber die Pyrazolgruppe " (1899) ; " Die
Halogenalkylate und quaternaren Ammoniumbasen " (1899) ;
" Die Nitrosoverbindungen " (1903) ; and " Ueber Chinone und
chinoide Verbindungen " (1906).

The relation between colour, fluorescence, and chemical
structure has been dealt with by Kauffmann in three pamphlets
belonging to the same series : " Ueber den Zusammenhang
zwischen Farbe und Konstitution bei chemischen Verbin-
dungen" (1904) ; " Die Beziehungen zwischen Fluoreszenz und
chemischer Konstitution" (1906); "Die Auxochrome" (1907) ;
but for a full account of these and allied subjects, Smiles'
volume, " The Eelations between Chemical Constitution and
some Physical Properties " (1910) should be consulted.

Finally, we may turn to those works in which the reactions
of the laboratory are classified and regarded from the point of
view of their value in synthetic work. Posner, in his " Synthe-
tischen Methoden der organischen Chemie," has inverted the
usual order of text-books ; for instead of classifying reactions
according to the compounds which they produce, he classifies
compounds according to the nature of the reaction which gives
rise to them. Lassar Cohn, in his " Allgemeine Gesichtspunkte
der organischen Chemie," has collected together many instances
in which reactions cannot be carried through conveniently in
the ordinary manner, and he shows how by resorting to various
artifices the required product may be obtained. The Grignard
reaction and its various applications have been described in detail
by Schmidt in " Die organischen Magnesium verbindungen und
ihre Anwendung zu Synthesen " (Ahrens Sammlung, 1905) ; and

296 RECENT ADVANCES IN ORGANIC CHEMISTRY

by McKenzie in the 1907 British Association Eeport on the
subject. The whole question of recent synthetic chemistry has
been dealt with by Schmidt in his work, " Die synthetisch-
organische Chemie der Neuzeit" (Die Wissenschaft, 1908).

In concluding this notice of the current literature of organic
chemistry, it may be well to point out one or two branches
which still remain unexhausted. In the first place, there is
room for a second work on the lines of Lachman's " Spirit of
Organic Chemistry," as a perusal of that volume generally
stimulates in the reader a desire for further essays of the same
type. Secondly, it seems possible that a very interesting
account of modern research in organic chemistry might be
written by taking up in turn the chief workers of the last two
decades and showing how from one investigation they were led
to another. Of course the Memorial Lectures of the Chemical
Society and the obituary notices of the deutschen chemischen
Gesellschaft cover some of this ground ; but in them we have
no account of many of the most important researches of the
last ten years, owing to the fact that the investigators are still
alive. Then, a third book might be written to show the
connection between organic chemistry and chemical industries.
To a great extent, the specialization in pure chemistry on the
organic side at the present day is caused by the difficulty which
the ordinary chemist encounters when he endeavours to get
into touch with the commercial side of the subject. A book
which would bridge this gap would be of considerable service.
Finally, one may suggest that a most useful volume might be
written by any one who would go to the original literature and
endeavour to deal critically with the constitutions attributed to
the commoner classes of organic substances. In comparing
text-books with original papers, one is often struck by the way
in which a tentative suggestion put forward in a paper becomes
dogmatic in tone when transferred to a text-book ; and any
one who would assess for us the value of the actual experi-
mental evidence in favour of, say, the presence of a carbonyl
radicle in a carboxyl group, would be doing work of much
interest and of very considerable importance.

INDEX OF NAMES

ACH, 161, 166
Ahrens, 161, 166
Anschiitz, 14, 291
Armitage, 290
Armstrong, 294
Arnold, 247
Arth, 62

Aschan, 34, 77, 293
Athanasescu, 145

Auwers, 207, 209, 211, 217, 218, 220,
235

BABO, 117

Baeyer, 5, 10, 11, 12, 26, 28, 35, 36, 60,

87, 88, 89

Baly, 215, 265, 267, 285
Bamberger, 201, 203, 205, 206, 209, 212,

213, 216, 219, 220, 285, 288
Bandow, 155
Barbier, 97, 98, 99, 102
Barthelt, 293
Bauer, 261, 262
Baumann, 211
Bayliss, 294
Becker, 148
Beckett, 146, 155
Beckmann, 62
Behrend, 161
Beilstein, 290
Senary, 190
Berkenheim, 63
Berthelot, 33
Bertrand, 247
Bischoff, 13, 291
Blaise, 44, 275, 276
Blanc, 72, 76, 79
Blangey, 206
Bode, 132

Borsum, 228, 230, 233, 235, 236
Bouchardat, 93

Bouveault, 84, 97, 98, 99, 101, 102
Boyd, 251
Brady, 213
Bredt, 76, 80, 81, 260
Brickner, 77
Brown, 131
Bruce, 27, 32, 34
Briihl, 32, 284

Bucherer, 295
Buchner, 25
Butlerow, 8

CAIN, 294, 295

Galdwell, 249

Caro, 206

Carroll, 161

Chick, 187

Claisen, 83

Cohen, 244, 291

Collie, 10, 38, 180, 190, 191, 192, 193,

194, 195, 196, 197, 281
Comstock, 135

Cone, 226, 230, 233, 234, 235, 238
Cooper, 247
Cotton, 251

Couper, 3, 17, 18, 282, 285
Crafts, 179
CrSpieux, 122
Curtius, 25, 177

DALTON, 16
Dewar, 14
Dobbie, 151
Dodge, 93
Drude, 15

EDWARDS, 267
Einhorn, 61, 127
Elkeles, 55
Erlenmeyer, jun., 247
Euler, 91

FALK, 288

Faraday, 288

Federer, 247

Feist, 190

Fenton, 241

Finkelstein, 145

Fischer, 13, 14, 111, 161, 162, 166, 169,

170, 171, 172, 173, 175, 177, 243, 234,

249, 294
Fisher, 41
Fittig, 21, 117, 119
Fliirscheim, 286, 288
Forster, 75, 267
Foster, 146

298

Frankel, 294
Frankforter, 157
Frankland, 3, 286, 287
Freer. 26

Freund, 26, 148, 157, 160
Fritsch, 159
Friedel, 179
Frohlich, 292

GAMS, 142

Gardner, 81

Gtenequand, 121

Gerhardt, 17

Gerngross, 169

Gildemeister, 55, 293

Ginsberg, 84, 85

Goeschen, 292, 295

Goldschmiedt, 140

Gomberg, 223, 224, 225, 226, 228, 229,

230, 231, 233, 234, 235, 237, 238, 239,

240

Goose, 292
Gourmand, 101
Grtebe, 5
Griess, 5

Grignard, 24, 40, 41, 44, 52, 206, 215
Guareschi, 111, 293
Guye, 15

HAAKH, 252

Haller, 61, 75, 76, 77, 79

Hantzsch, 10, 13, 119, 120, 292, 294

Harden, 294

Hardy, 294

Harries, 69, 94, 102, 260, 275

Hartley, 15

Haworth, 28

Hayakawa, 255

Heintschel, 232, 236

Henle, 252

Henrich, 291

Hentzschel, 21, 28

Hermann, 10

Herzig, 114

Hesse, 61

Heusler, 292

Hilditch, 191, 290

Himmelmann, 102

Hinrichsen, 261

Hirschberger, 243

Hjelt, 295

Hoff, van't, 12, 15, 16, 292

Hoffmann, 293

Hofmann, 153

Homfray, 193

Hopkins, 294

Horbaczewski, 161

Huber, 121

IGLAUEE, 128
Ipatjeff, 34, 91, 97

INDEX OF NAMES

JABLONSKI, 275

Jackson, 76

Jacobson, 10, 230, 232, 233, 235, 236,

291

Jahns, 119, 120
Johnson, 69
Jones, 247, 292
Jiinger, 63

KACHLER, 78

Kallen, 260

Kametaka, 29

Kauffmann, 285, 295

Kay, 52

Keane, 291

Kehrmann, 215, 231

Kekule, 3, 4, 5, 6, 8, 11, 12, 14, 15, 17,

260, 282, 285
Keller, 117
Kerscnbaum, 102
Kipping, 13, 24, 247, 251
Kishner, 22
Klages, 61, 63, 263
Kleber, 32
Klever, 186
Komppa, 72
Kondakow, 69

Konigs, 134, 135, 136, 137, 138, 140, 169
Koppe, 118
Ko'rner, 5, 14
Koster, 111
Kotz, 91, 62
Kriigel, 27
Kriiger, 104

LAAE, 8, 9, 10, 264

Labbe, 95

Lachman, 291, 296

Ladenburg, 5, 6, 27, 115, 116, 119, 130,

131, 290, 292
Laiblin, 121
Landolt, 292
Lapworth, 80, 293
Lassar-Cohn, 290, 295
Lauder, 151
Lawrence, 44
Leathes, 294
Le Bel, 12, 13, 15, 16
Lees, 34
Lellmann, 275
Lenton, 80
Leser, 97
Levallois, 84
Ling, 294
Lippmann, 294
Loew, 241
Lohse, 261
Lowry, 293
Luniak, 169

INDEX OF NAMES

299

MALIN, 78

Mamlosk, 292

Mann, 294

Maquenne, 294

Marchlewski, 111

Marckwald, 244, 249, 250, 292

Markownikoff, 28, 29, 35, 59, 254

Marsden, 267

Marsh, 81

Martine, 61

Matthiessen, 146

MoKenzie, 244, 245, 249, 250, 296

Mehrlander, 62

Meldola, 293

Mencke, 76

Meth, 250

Meyer, E., 290

Meyer, H., 114

Meyer, J., 251

Meyer, R., 291

Meyer, Victor, 6, 13, 291

Meyerhoffer, 292

Michael, 253, 255

Miller, 137, 138

Montgolfier, 77

Morgan, 294

Moser, 24, 262

Miiller, 266

Muir, 290

Mumme, 255

Myers, 196

NEF, 261
Nelson, 288
Nencki, 111
Neuberg, 247
Neville, 13
Nietzki, 294
Norris, 231

OPPEBMANN, 276
Osborne, 294
Ostromisslensky, 247
Ostwald, 282

PASTEUR, 1, 12, 246, 247, 248, 251

Paternd, 12

Paul, 250

Peachey, 13

Perkin, Sir W. H., 16, 119, 291

Perkin, W. H., 24, 26, 28, 34, 39, 41,52,

57, 60, 72, 74, 80, 157
Pictet, 111, 121, 122, 123, 142, 145, 293
Pinner, 121
Plimmer, 294
Pope, 13, 247
Posner, 276, 295
Pummerer, 193, 197

RAMSAY, 2, 19, 115
Rayleigh, 19

Remsen, 117, 119
Reyher, 80
Riban, 77

Richter, M. M., 290
Richter, V., 14, 291
Roberts, 271
Robinson, 157, 230
Rohde, 137
Roosen, 161
Roscoe, 291
Rosenberg, 76
Roser, 147, 157
Rudolf, 212
Riigheimer, 119, 130
Rupe, 295

SABATIEE, 22, 28, 30, 34

Salway, 151

Sanders, 231

Schauwecker, 94

Schindelmeister, 69

Schmidlin, 238

Schmidt, 42, 93, 94, 95, 99, 104, 291,

292, 293, 295, 296
Scholtz, 119, 246, 292, 293
Schorlemmer, 290, 291
Schotten, 211
Schroder, 285
Schryver, 294
Schulz, 294
Schulze, 169
Schwarz, 61, 62

Semmler, 69, 84, 90, 99, 102, 293
Senderens, 22, 28, 30, 34
Simonsen, 44
Skraup, 136, 139
Slawinski, 85
Slimmer, 244

Smiles, 13, 32, 37, 246, 287, 295
Sobrero, 85
Staudinger, 186, 187
Stewart, 186, 194, 196, 215, 263, 265,

267, 285
Stohmann, 32
Strecker, 171
Sudborough, 271

TATTEBSALL, 57

Thiele, 16, 45, 51, 258, 259, 260, 268,

274, 279
Thomas, 271
Thompson, 250
Thorpe, 72, 74, 80, 295
Tickle, 192
Tiemann, 42, 93, 94, 95, 97, 98, 93, 100,

101, 102, 104, 118
Tigges, 99
Tilden, 93
Tinkler, 151
ToUens, 294
Tolloczko, 247

INDEX OF NAMES

Traube, 161, 162

Tschitschibabin, 229, 230, 235, 236
Tschugaeff, 90

ULLMANN, 228, 230, 233, 235, 236

VEEAGUTH, 29

Verley, 97

Vorlander, 255, 256, 257, 271, 276, 279

WAGNEE, 41, 64, 77, 81, 85

Walden, 225, 238, 292

Walker, 131

Wallach, 11, 42, 43, 50, 55, 66, 68, 76,

82, 85, 293

Wedekind, E., 246, 247, 251, 292, 295
Wedekind, 0., 246, 251
Wegscheider, 118
Weidel, 121

Werner, 13, 16, 286, 292
Weyl, 290

Whitely, 244

Wiederhold, 207

Willstatter, 27, 29, 32, 34, 124, 127, 128,

130, 132, 193, 197, 266
Wilsmore, 186, 187, 188, 189
Wislicenus, J., 12, 21, 28
Wislicemis, W., 293
Wohler, 147, 150
Wolffenstein, 155
Wren, 245
Wright, 146, 155
Wurtz, 21

YATES, 34
Young, 28

ZALESKI, 111

Zeisel, 113, 114

Zeitschel, 102, 107

Zelinsky, 24, 30, 52

Zincke, 202, 203, 204, 207, 216, 218, 229

INDEX

SUBJECTS

ABSOEPTION spectra, 15, 265, 266, 271, 285
Acetalamine, 159
Acetaldehyde, 118
Acetamide, 187
Acetanilide, 187

Acetic acid, 115, 180, 186, 187, 197
Acetic anhydrides, 186, 191
Acetic ester. See Ethylacetate
Acetoacetic acid, 180, 186, 197
„ anilide, 187

ester, 9, 80, 189, 197, 222, 260, 263, 266, 278
,, ,, structure of, 9

,, ,, condensation in ring syntheses, 25, 26, 61

Acetobutyl hydrazone-hydrazide, 187

,, iodide, 24
Acetone, 115, 187, 197, 263
,, dicarboxylic ester, 263
„ relations to keten group, 197
Acetonyl -acetone, 263
Acetophenone, 130, 132
Acetophenone chloride, 130
Acetoveratrone, 143
Acetylacetone, 97, 197, 263

,, from keten, 197

Acetyladipic ester, 44
Acetylation of quinoles, 217, 220
Acetyl chloride, 132, 143, 187, 197, 211
Acetyl bromide, 211
Acetylene, 116, 132, 270

„ compounds, 270
Acetylglutaric ester, 44
Acetylketen. See Diketen
Acetylsuccinic ester, 44
Acidic hydrogen and unsaturation, 278
Acrylic acid, 261
Active solvent, action of, 247
Addition reactions, 253 ff.

,, „ definition of, 253

„ „ degrees in, 263

Adipic ester, 25
Affinity rearrangement, 264 ff.
Alanine, 171

,, leucine, 171
Albumins, 169, 170, 176

„ hydrolysis of, 170
„ properties of, 169, 176
Albumoses, 170

302 INDEX OF SUBJECTS

Alkaloid constitutions, 113 ft.

„ „ methods of determining, 113

Alkaloids, 14, 110 ff.

,, classes of, 112
,, decompositions of, 113
„ definition of, 110
„ extraction of, 112

general properties of, 112
isoquinoline group of, 140 ff.
morpholine or phenanthrene group of, 112
occurrence of, 111
practical value of, 110
production of, in nature, 111
purine group of, 161 ff .
pyridine group of, 115 ff.
pyrrolidine group of, 120 ff.
,, quinoline group of, 134 ff.
,, synthesis, 115 ff.

„ „ examples of complete, 115 ff., 131

Allantoin, 165
Alloxan, 164
Allyl alcohol, 115
,, bromide, 115
„ methyl ketone, 273
Aluminium amalgam, 245
„ chloride, 178
Amido-acids from proteins, 170
„ resolution of, 171
Amidomalonyl urea, 162
Amidopyridine, 122
Amido-tetramethylene, 28

„ -uracil, 163

Amino-aceto-veratrone, 142, 143
Ammonia's action on unsaturated ketones, 275 ff.
Amyl nitrite, 143
Anhydrocamphoronic acid, 80
Aniline, 82, 187
Apocamphoric acid, 81, 82, 83
Apophyllenic acid, 149, 150
Arabite, 248

Aromatic compounds, Kekule's views on, 3
Artificial camphor, 87
Asymmetric amide formation, 250
„ carbon atom, 12, 15

esterification, 249
hydrolysis, 249
syntheses, 108, 241 ff.

„ definition of, 243

„ failure of, in cases of nitrogen and sulphur, 246
Atrolactinic ethyl ether nitrile, 130
Atropic acid, 130
Atropine, 111, 131

,, constitution, 131
„ synthesis, 131
Azo-compounds from quinoles, 220

BACTERIUM xylinum, 247
Baeyer's Strain Theory, 35, 288
Barbituric acid, 162
Baumann-Schotten reaction, 211
Benzaldehyde, 130, 146, 251

INDEX OF SUBJECTS 303

Benzene, 3, 4, 5, 6, 7, 8, 9, 10, 11, 14, 28, 132, 180, 193, 194, 260
,, addition of bromine to, 271
,, Baeyer's researches on, 10

„ derivative from diacetylacetone, 183, 184, 186, 196
„ formulae, 6 ff.
,, Kekule on, 3
,, orientation of substituents, 4
„ Thiele's formula for, 260
Benzenoid and quinonoid character, 215, 217
Benzoic acid, 13, 146
Benzoin, 251

Benzoylformic acid, asymmetric synthesis of mandelic acid from, 245, 251
Benzpyrene compound, 198
Benzyl alcohol, 146
Benzyl-hydrocotarnine, 154
Bertrand's method, 247
Bibliography, 290 ff.
Biochemical methods of resolution, 247
Bishydroxymethylene-acetone derivatives, 197, 199
Biuret reaction, 176

Boiling-points of paraffins, defines, and polymethylenes, 31
Bonds, conjugated double, 259
Bonds, variation in strength of, 287
Books, annual summaries, 291

monographs, 291, 292, 294, 295
of essays, 291
on alkaloids, 293
, biochemistry, 294
, diazo compounds, 294
, dyestuffs, 294
, history, 290
, practical details, 290
, purines, 294
, reactions, 295

, relations between constitution and physical properties, 295
, stereochemistry, 291
, sugars, 294

, synthetic chemistry, 295
, terpenes, 293
reference, 290
some suggested, 296
text-books, 291
Borneol, 76 ff.

,, constitution of, 76
„ in asymmetric syntheses, 246
Bornyl bromide, 77
,, chloride, 77
„ iodide, 77, 89
Bornylene, 77, 89

,, constitution of, 89
Bromacetic ester, 80
Bromacetylbromide, 186, 187
Bromacrylic acids, 261

Bromine, 125, 125, 187, 207, 257, 261, 262, 269, 271, 286
,, addition reactions, 33, 261 ff.
,, atom, its effect on additive power, 261
Bromoacetoacetic ester, 187
Bromoacetoacelyl bromide, 188
Bromoanhydrocamphoronic chlorides, 81
Bromocampholic acid, 79
Bromocamphoric acid, 74

304 INDEX OF SUBJECTS

Bromooamphoric anhydride, 80
Bromohexahydrotoluic acid, 57
Bromoisobutyric ester, 66, 80
Bromo-isocaproyl-diglycylglycine, 175, 176
Bromomaleic acid, 262
Bromomesaconic acid, 262
Bromostilbene, 262
Bromotropane, 127, 130

„ methyl ammonium bromide, 126

Bromotropidine hydrobromide, 127
Brucine, 244
Butane, 81, 32
Butylene, 31, 32
Butyraldehyde, 189

CAFFEINE, 166, 168

„ synthesis, 166

Calcium salt method of preparing polymethylenes, 21
Camphane, constitution of, 76 ft.
Camphanic acid, 79, 80

,, ,, synthesis of, 80

Camphene, 72, 76 ff., 81, 85
Camphlene glycol, 81
Camphoic acid, constitution of, 81
Campholic acid, 78
Campholide, 75, 79
Camphor, 11, 72, 75, 76, 78, 83, 109

artificial, 87

„ as active solvent, 251
,, synthesis of, 75 ff.
Camphoranic acid, 81
Camphoric acid, 11, 72, 74, 75, 79, 83, 89

„ „ racemization phenomena of, 74

„ „ synthesis of, 72 ff.

„ acids as asymmetric hydrolyzing agents, 249
Camphoric anhydride, 74, 75

„ „ action with aluminum chloride, 34

Camphoronic acid, 79, 80, 81

„ „ constitution of, 80

,, ,, synthesis of, 80

Camphor quinone, 268

Camphorsulphonic acids as asymmetric hydrolyzing agents, 249
Cane-sugar, attempted asymmetric hydrolyses of, 249
Carbethoxy-glycylglycyl-leucine ester, 173
Carbocinchomeronic acid, 141
Carbohydrates, 169
Carbon atom, asymmetric, 12

„ dioxide, reduction to formaldehyde, 241
„ disulphide, 115-224
„ oxysulphide, 90
,, tetrachloride, 115
Carbonium bond, 285
Caro's reagent, 206
Carone, 59, 60

„ oxime, 60
Caronic acid, 59
Carvenone, 109
Carvestrene, 56 ff .
Carvomenthol, 70
Carvone, 47, 58, 69, 70
„ hydrobromide, 69

INDEX OF SUBJECTS 305

Carvoxime, 46

Carylamine, 60

Gatechol, 202

Chloracetal, 159

Chloracetyl chloride, 174

Chlorine, its action on hexamethylene, 34

„ ,, phenols, 202

Ghloro-carbonic ester, 173
Chlorocymene, 63
Chloroform, 112, 224, 225
Chloro-formic ester, 164
Chlorophyll, 111
Chlorotheobromine, 166
Chlorotheophylline, 165
Chromic acid, 115, 121, 134, 135, 139, 202
Cinchene, 135
Cincholoiponic acid, 136
Cinchonic acid, 134, 139
Cinchonidine, 140
Cinchonine, 134, 135, 139, 140

constitution, 134 ff.
„ " second half " of, 134 ff .

Cinchotenine, 135
Cinchotoxine, 138
Cineol, 52 ff., 85

„ oxidation products of, 55
Cineolic acid, 55, 56

„ anhydride, 56
Cinnamenyl-cyanacrylic ester, 261
Cinnamylidene-malonic acid, 261
Circularly polarized light, 251
Cis-terpin, 53, 54, 55

„ dehydration products of, 54
„ dibromide, 53, 54, 55
„ its conversion into trans-terpin, 53
Citral, 96, 100, 102 ff., 106, 107
„ group, 96 ff.

,, its conversion to cyclocitral, 103, 104
,» ), „ cymene, 102

,, stereoisomerism, 102, 107
„ synthesis of, 102
Citronellal, 93, 94, 95, 96, 100, 101, 102
„ dimethyl-acetal, 94
,, constitution of, 93
Citronella oil, 93
Citronellic acid, 93, 100
Citronellol, 93, 101
Claisen's reaction, 118
Cocaine, 111, 133

„ synthesis, 133
Collie's formula for benzene, 10

,, „ „ dehydracetic acid, 189

Complete alkaloid synthesis, examples of, 115, 131
Conchinine, 140

Condensation methods of synthesizing polymethylenes, 25, 26
Conductivity, electrical, 15, 225, 263, 271
Coniine, 115 ff., 125

„ synthesis, 115 ff.
Conjugated double bonds, 51, 259
Constitution and reactivity, 261 ff.

„ determination in alkaloid series, 113 ff.

306 INDEX OF SUBJECTS

Constitution determination in albumin group, 170
Cotarnic acid, 147, 148
Cotarnine, 147, 149, 150, 154, 160

„ constitution, 147 ff.

,, solvents' effects on, 151

,, synthesis of, 151 ff.

„ tautomeric forms of, 150
Cotarnolactone, 147, 149
Cotarnomethine methyl iodide, 147, 149
Cotarnone, 147, 149
Couper's formulae, 3

„ views on formulae, 17, 282
Cresol, 203, 207, 208
Cresorcinol, 214
Crotonic acids, 6, 261
Cryptogamia, 111
Cyanacetic acid, 163

ester, 39, 104
Cyanacetyl-urea, 163
Cyanhydrin reaction, 171, 243
Cyanopentane tricarboxylic ester, 39
Gyclobutene, 21, 28
Cyclocitral, 103, 104
Cyclogeranic acid, 99
Cycloheptadiene, 125
Cycloheptatriene, 124, 125
Cycloheptene, 124, 125
Cyclohexanone carboxylic ester, 52
Cyclo-octadienes, 30
Cycloparaffins. See Polymethylenes
Cyclopentadiene, 21
Cyclopentanone, 21
Cyclopropane, 21
Cymene, 102, 103

DBHYDBACETIC acid, 180, 182, 186, 189, 190, 191
,, „ constitution of 189 ff.

isomer of, 186, 191

„ „ table of its derivatives, 186

Dehydration reactions, 44, 50, 54, 64, 66
Dehydrocamphoric acid, 73, 74
Desmotropic bodies, reactivity of, 179

„ „ magnetic rotation of, 15

Desmotropy, 10
Desoxymesityl oxide, 273
Diacetonamine, 273
Diacetone hydroxylamine, 67
Diacetylacetone, 182, 183, 186, 195, 197

„ derivatives. See Benzene, Dimethylpyrone, Isoquinoline,

Lutidene, and Naphthalene
,, properties of, 195

Diacetyl-dimethylpyrone, 195

,, -hydroxy-benzyl-alcohol, 220
Diacetylorcinol, 195
Diacetylpentane, 24
Diamidouracil, 164

,, ur ethane, 164

Diazo-compounds, isomerism of, 13

,, „ used in polymethylene syntheses, 25

Diazo-m ethane, 25

INDEX OF SUBJECTS 307

Dibenzal-tropinone, 130

Dibromocotinine, 121, 122

Dibromocrotonic acid, 262

Dibromofumaric acid, 262

Dibromohexamethylene, 28

Dibromostilbene, 202

Dibromoticonine, 121, 122

Dichloropropylene, 27

Dichlorotetrahydrocymene, 63

Dielectric constant of solvent in addition reactions, 257

Dihydrocarveol, 47, 70

Dihydrocarvone, 58

,-, hydrobromide, 58

Dihydronicotyrine, 123
Dihydroxycamphoric acid, 73
Dihydroxydihydrocitronellal acetal, 94
Dihydroxylene, 98
Dihydroxy-nonane, 24
Dihydroxyphenyl-acetic acid, 182
Di-iodopurine, 167
Di-isonitroso-tropinone, 130
Di-isoprene, 93

Diketen, 180, 186, 187, 189, 197
„ properties of, 187 ff.:
„ refractivity of, 188
Diketides, 180, 189
Diketoapocamphoric acid, 82
„ ester, 72

Diketocamphoric ester, 72
Diketohexamethylene, 28
Diketopiperazine, 175
Dimethoxy-quinoline, 140
Dimethyl-acrylic-acid, 261
,, allene, 92
„ amine, 125
,, aminocycloheptadiene, 126
,, dihydroxy-heptamethylene, 24
,, fumaric acid, 262
,, glutaric ester, 72
,, granatanine, 29
„ hydroxylamine, 206
pentamethylene, 35
homocatechol, 141, 142
methylpyrrolidinium iodide, 92
methylpyrrolidine, 92
pentamethylene, 35
pyrone, 182, 186, 191, 192, 193, 194, 195, 196, 270

constitution of, 193
,, isomer, 194
„ properties of, 193
„ salts, 192
quinole, 206, 207
sulphide, 270

trimethylene dicarboxylic acid, 59
urea, 165

uric acids, 165, 166
Dinaphthol derivative of diacetylacetone, 198
Dipentene, 44 ff., 50, 55, 93

,, action of nitrosyl chloride on, 45

,, constitution of, 44 ff .

„ hydrobromide, 53, 54, 55

3o8 INDEX OF SUBJECTS

Dipentene nitrosochloride, 46
Diphenyl-bromo-ethylene, 262
» ,, propylene, 262

ethylene, 262, 286
,, -methyl-propylene, 262
„ propylene, 262

Dissociation constant and additive power, 263
Double bonds, action of nitrosyl chloride on, 45
„ and hydroxylamine, 67
,, and niercaptans, 276
„ conjugated, 51, 259
„ their behaviour on oxidation, 40
salts, 270, 272
Dynamic and static views of chemistry, 16
,, views of addition reactions, 263

ECGONINE, 132, 133

,, constitution, 132 fi.

„ synthesis, 132 ff.

Electric absorption, 15
Electrical conductivity, 15, 225, 263, 271
Electric discharge, silent, 38
Electrons, 288, 289
Esterification; selective partial, 246

,, unsaturation's influence on, 271

Ethane tetracarboxylic ester, 26, 27
Ethereal oils, 38, 91
Ethyl acetate, 152, 187, 189, 197
Ethylene, 36, 39, 132, 269, 286

Ethylenes and polymethylenes, relative degrees of unsaturation, 33
Ethylidene-propionic acid, 202
Ethyl-sulphuric acid, 132
Exhaustive methylation, 29

FAEADAY tube, 288

Fats, 169

Feist's formula for dehydracetic acid, 190

Fenchane, constitution of, 84

Fenchene, 82, 83, 85

,, constitution of, 82 ff.
Fenchocamphorone, 82, 83
Fenchone, 82, 84

,, constitution of, 82 fi.
Fenchyl alcohol, 82, 84

„ „ constitution of, 84

„ chloride, 82
Ferments, 170, 177

Fischer's methods of synthesizing polypeptides, 172 ff.
Fittig-Wiirtz reaction, preparation of polymethylenes by, 21
Formaldehyde, 111, 178, 241

,, conversion of, into fructose, 141

Formic acid, 68
Formulae, views of Couper and Kekule on, 17, 18

„ implications in modern, 283
Friedel-Crafts' reaction, 132, 179
Fructose from formaldehyde, 241

„ syntheses in plants, 241
Fulvene, 287
Fumaric acid, 262

„ f ester, 25
Fungi, resolution of racemates by, 247

INDEX OF SUBJECTS 3<>9

GALLIC acid, 147, 148
Geranial, 107
Geranic acid, 96, 98 ff.

„ ,, synthesis of, 98
Geraniol, 96, 102, 106 ff.
Glucose, 111, 251
Glutaric acid, 131

,, nitrile, 115
Glycerine, 115, 131
Glycine, 173, 174, 175

,, ester, 174
Glycocoll, 173
Glycyl, 173

,, chloride, 175
Glycylglycine, 173, 174, 175

,, carboxylic ester, 173

ester, 173

„ glycine, 175

„ leucine, 173

Glycylglycylglycylglycine carbethoxy-ester, 173
Glycylglycylleucine carboxylic ester, 173
Gnoscopine synthesis, 157
Grignard reaction in terpene syntheses, 40, 41, 44, 52, 57

„ „ „ quinole syntheses, 206, 215

Guye's hypothesis, 15

HAEMOGLOBIN, 111

Hantzsch- Werner theory, 13

Heat of combustion of polymethylenes, 32

,, formation of trimethylene, 33

„ bromine addition to ,, 33

„ sulphuric acid ,, ,, 33
Helicin, 244
Hemipinic acid, 147
Hemiterpenes, 39
Heptamethylene, 29, 31, 35, 36

„ method of preparing, 29

Heptane, 31, 32
Heptene, 31, 32

Herzig and Meyer's method, 114

Hexachloro-diketo-K-hexen, 202

„ E-penten, 202

„ „ -hydroxy-carboxylic acid, 202

Hexahydrocinchomeronic acid, 136
Hexahydrocymene, 39
Hexahydro-hydroxy benzoic acid, 57
Hexahydroxylylic acid, 34
Hexamethylene, 28, 31, 33, 34, 35, 36

„ conversion into methyl-pentamethylene, 34

„ methods of preparing, 28

Hexane, 31, 32
Hexaphenyl-ethane, 223
Hexene, 31, 32

Hindrances to reactions, 13, 261 ff .
Hofmann's reaction, 153
Homocamphoric acid, 76
Homophthalic acid, 76
Homoprotocatechuic acid, 143
Homoterpenylic acid, synthesis of, 44

,, „ methyl ketone, 42

Homoveratric acid, 143

3io INDEX OF SUBJECTS

Homoveratroyl chloride, 143

„ -amino-aoeto-veratrone, 144

-hydroxy-homoveratrylamine, 144
Hydrastine, 158, 159, 161

,, constitution, 160

„ hydrochloride, 160

„ synthesis, 158

Hydrazone formation applied to resolution of racemates, 247
Hydriodic acid, 34, 35, 129, 255

Hydrobromic acid, 40, 53, 55, 57, 59, 69, 115, 125, 127, 209, 210, 211, 216, 262
Hydrochloratropic acid, 131
Hydrochlorocarvoxime, 87
Hydrocotarnine, 146, 155, 156
,, synthesis, 155

Hydrogen peroxide, 115
Hydrohydrastinine, 159, 160
Hydroquinone, 201
Hydroxybenzoic acid, 57
Hydroxycamphoronic acid, 81
Hydroxydichloropurine, 167
Hydroxydihydrogeranic acid, 99
Hydroxyfenchenic acid, 82, 83
Hydroxyhexahydrotoluic acid, lactone of, 57

ester, 40
Hydroxylamine, 193, 212, 213, 275

„ and double bonds, 67, 275

Hydroxymenthylic acid, 62, 65
Hydroxytoluic acid, 48
Hydroxytrimethylglutaric ester, 80

IMPLICATIONS in modern formulae, 283
Indiarubber, isoprene produced from, 92
Influence of constitution on reactivity, 261 ff .
Intramolecular acetoacetic ester condensation, 25, 26, 61

„ changes in quinoles, 215 ff.

Iodine bromide, 255

„ chloride, 115, 255
lodoacetic ester, 98
lodoheptamethylene, 35
lodopropionic ester, 39
Ionic hypothesis, 18
lonone, 104, 105
Irone, 106
Isoborneol, 77
Isocamphoric acid, 75
Isoconiine, 116
Isocrotonic acid, 261

Isomeric form of dimethyl-pyrone, 194. See also Orcinol
Isomyristicin, 151, 152
Isophenylcrotonic acids, 276
Isoprene, 91, 92

,, polymerization of, 92, 93
„ syntheses of, 91
Isopropyl alcohol, 115
Isopulegol, 95
Isoquinoline alkaloids, 140 ff.

„ derivative of diacetylacetone, 183, 185, 186, 196

KEKULE'S formulae, 3

,, vibration hypothesis, 6 ff.

„ views on formulae, 17, 282

Keten, 179, 180, 186, 187, 189, 197

INDEX OF SUBJECTS 31 1

Keten properties of, 186

„ table of its derivatives, 186
Ketobromides, 208
" Keto-enol " type, 179
Ketohexahydrobenzoic acid, 39, 57
Ketopentamethylene, 21, 22

M „ carboxylic ester, 25

Kishner's method, 22
Konigs' cinchonine formula, 138, 139

„ loiponic acid formula, 137

LAAR'S hypothesis, 9, 264
Labile groupings of atoms, 8 ff.
Lactic acid, 245, 251
„ ,, asymmetric syntheses of, 245
„ isomerism, Wislicenus on, 12
Lactones, formation of, 283, 284
Lsevulinic acid, 97, 106

ester, 263

Laudanosine, synthesis, 145
Le Bel-van't Hoff theory, 12
Lepidine, 135, 136
Leucine ester, 173
Leucyl-octaglycyl-glycine, 176
Leucyl-triglycyl-leucyl-octaglycyl-glycine, 176
Leucyl-triglycyl-leucyl-triglycyl-leucyl-octaglycylglycine, 176

Limonenes, 45, 50

„ constitution of, 50

Linalool, 96, 106, 107, 108
Loiponic acid, 136, 137
Lutidone, 183, 186, 196

MAGNESIUM methyl iodide, 207
Magnetic fields, 251

„ rotatory power, 15, 271, 285
Maleic acid, 262
Malonic acid, 121, 162

„ ester polymethylene syntheses, 25, 26
Malonylurea, 162

Mandelic acid, asymmetric esterification of, 249
„ ,, ,, synthesis of, 245

,, amides, 250

„ esters, hydrolysis of, 250

„ ,, selective formation of, 249

Mannose in cyanhydrin reaction, 243
Markownikoff Rule, 59, 255
Meconine, 146, 156
Menthene, 63 ft.

„ tertiary, 70
Menthenone, 69
Menthol, 63 ff.

„ in asymmetric syntheses, 244, 245

„ in selective esterification and hydrolysis, 249
Menthone, 61 ff.

,, decomposition products of, 62
,, from rhodinal, 101
Menthylamine, 250
Menthyl mesaconate, 244
Mercaptans and double bonds, 276
Meroquinene, 134, 135, 136, 137, 139
Mesaconic acid, 244, 262

312 INDEX OF SUBJECTS

Mesaconic menthyl ester, 244
Mesityl oxide, 67, 273, 274, 275

,, oxime, 67
Mesoxalic acid, 164
Mesoxalyl-urea, 164
Methoxyl groups, estimation of, 113
Methyladipic acid, 62, 65, 93
Methylamine, 121
Methylation, exhaustive, 114
Methylcyclohexanone, 66, 68
Methyl dibromobutane, 92, 97
Methylenedihydroxy-isoquinoline, 159
Methyl-ethyl-malonic acid, 244

„ granatanine, 29

„ heptenone, 56, 96, 97, 98, 102, 106

,, ,, constitution of, 97

,, ,, syntheses of, 96 ff.

„ hexamethylene, 35

,, imino groups, determination of, 114

,, isopropyl-pimelic acid, 61

„ mercaptan, 90

„ methoxy-hydrastinine, 160

,, „ methylenedioxy-benzyl-dihydro-isoquinoline, 154

„ ,, methylene-dioxy-phenylpropionic acid, 151, 152

„ methylene-gallic-carboxylic acid. See Gallic acid

,, pentamethylene, 24

„ phenyl-glycollic acid, asymmetric synthesis of, 245

„ pyrrolidine, 92

„ quinole, 210, 214

„ radicle, its effect on additive power, 261

„ stilbene, 262

„ succinic menthyl ester, 244

„ tetrahydropapaverine, 145

„ tropane, 126

,, tropidine, 126
Michael's Distribution Principle, 255

„ theory, 253
Miller and Rohde's cinchonine formula, 138, 139

,, ,, loiponic acid formula, 137

Modifications of Pasteur's resolution methods, 247
Molecular refraction and ring formation, 32

,, volumes, 32
Monobromacetic ester, 80
Monocotyledons, 111
Monoketides, 180
Morpholine alkaloids, 112
Mucic acid, 122
Myristicin, 151, 152

„ aldehyde, 151, 152

NAPHTHALENE derivative of diacetylacetone, 183, 184, 186, 196
Narceme constitution, 157
Narcotine, 146, 147, 156, 157, 158, 161

„ constitution, 156, 157

„ synthesis, 157

Nef's views, 261
Neral, 107

Nerol, 96, 102, 106 ff .
Neutralization of affinity, 253
Nicotine, 120 ff .

„ constitution, 120 ff.

„ synthesis, 122

INDEX OF SUBJECTS 313

Nicotinic acid, 119, 120, 121
Nicotyrine, 123

„ synthesis, 123

Nitration, its influence on addition reactions, 262
Nitrogen compounds, optically active, 13

,, series, attempts at asymmetric synthesis in, 246
Nitromethane, 278
Nitro-phenylhydrazine, 212
Nitroso and iso-nitroso derivatives, 45
Nitrosyl chloride, 45, 46, 48, 87

„ „ action of, on double bonds, 45

Nonane, 31, 32
Nonomethylene, 30, 31, 32, 35, 36

,, preparation of, 30

Nonylene, 31, 32
Normal phellandrene, 69

OCTADECAPEPTIDE, 175

Octane, 31, 32
Octene, 31, 32
Octomethylene, 29, 30, 31, 35, 36

„ method of preparing, 29 ff.

Olefmes, boiling-points of, 31, 32

„ molecular volumes of, 32
Olefinic terpenes, 38, 91 ff .

,, ,, importance of, 91

" Onium " salts, 272
Opianic acid, 146, 147, 156, 160
OpticaUy active bodies, 12 ff.

„ „ new methods of producing, 246 ff.

„ solvent, 247, 250

Optical rotatory power, 15
Orcinol, 182, 183, 186, 194, 196
Orientation in benzene series, 4
Oscillation hypothesis of Kekule, 6 ff.
Oxalic acid, 81, 106, 121, 162

„ ester condensation applied to ring syntheses, 25, 73
Oxanthranol, 201
Oxenes, 193

Oximes, isomerism of, 13
Oximidomalonylurea, 162
Oximino-acids, 276
Oxonic acid, 165
Oxonium salts, 192, 193
Oxygen, quadrivalent, 192, 193
Oxymethylene derivatives, formation of, 83

PAPAVBEINB, 140, 141, 142, 144, 145

,, constitution, 140 ff.

,, chloro-methyl derivative of, 145

„ synthesis of, 142 ff.

Papaveroline, 140
Paraffins, boiling-points of, 31, 32

,, molecular volumes of, 32
Paraldehyde, 116
Partial valencies, 258

definition of, 258

Pasteur's methods of resolution, 246
Pelletierin. See Pseudo-pelletierin
Pentabromo-toluquinole, 208
Pentaglycylglycine, 176

3 [4 INDEX OF SUBJECTS

Pentamethylene, 22, 28, 31, 32, 33, 34, 35, 36

„ ' method of preparing, 22 .
Pentamethylene diamine, 115, 117
Pentane, 31, 32
Pentaphenyl-ethane, 230
Pentene, 31, 32
Peptones, 14, 170

„ decomposition products of, 14, 170
Perchlorethylene, 115
Perchloro- vinyl-acetic acid, 202
Perkin's reaction, 119

Permanganate. See Potassium permanganate
Phellandrene, 68

„ constitutions, 68 ff.

Phenanthrene alkaloids, 112

Phenyl-acetyl-methoxy-methylenedioxy-phenyl-ethylamine, 153
Phenylhydrazine, 14, 187, 190, 193, 212, 213, 220
Phenyl-methyl-pyrazyl-phenyl-methyl-pyrazolone, 190
Phenyl radicle, its action on double bonds, 262
Phosphoric acid, 135
Phosphorus pentabromide, 209, 210

„ pentachloride, 135, 143, 174, 190, 230
„ oxychloride, 163, 165, 166

trichloride, 165, 167
Phosphotungstic acid, 176
Phthalic acid, 75
Phthalide, 75, 76
Physical methods in organic chemistry, 15, 18

„ properties and chemical constitution, 284

„ „ and reactivity, 285

„ „ of saturated and unsaturated bodies, 76 ff., 271

Picoline, 116
Pinacoline, 263

Pinacone formation applied to polymethylene syntheses, 24
Pinene, 85, 87, 88, 89

,, constitution of, 85 ff.
„ hydrochloride, 87
,, nitrosochloride, 87
Pinic acid, 88

„ constitution of, 88
Pinol, 85, 86, 87

,, constitution of, 85 ff.
Pinolglycol, 85
Pinonic acid, 88, 89
Piperic acid, 117, 118, 119
Piperidine, 115, 117, 119
Piperine, 117, 119

„ constitution, 117 ff.

„ synthesis, 117 ff.
Piperonal, 117, 118, 159
Piperonalacetalamine, 159
Piperonylacrolein, 118
Piperonylic acid, 117, 118
Piperyl chloride, 119
Plant syntheses, 241, 252
Plants and animals, 178
Polarized light, absorption of, 251
Polyketides, 178 ff.

,, chief properties of class, 200
„ explanation of name, 180
„ table of, 197, 198
Polymerization of ethylene, 38

INDEX OF SUBJECTS 315

Polymerization of keten, 179
Polymethylenes, 20 ff., 270

and ethylenes, relative degrees of unsaturation, 33, 272
boiling-points of, 31, 32
,, chemical properties of, 33

heats of combustion of, 32, 33
,, methods of synthesizing, 21 ff.

,, molecular volumes of, 32

,, nomenclatures of, 20

„ refractivities of, 32

,, stabilities of, 33

Polypeptides, 169 ff.

definition of, 171
,, properties of, 176

Polypeptide syntheses, 172 ff.
Potassium cyanate, 162
,, cyanide, 251

permanganate, oxidation with, 33, 41, 42, 43, 55, 62, 64, 81, 82, 85,

86, 94, 115, 117, 120, 135, 136, 141, 147, 151, 165, 272
,, sulphate, dehydration by, 44, 50, 57
Potential differences, 257
Propane, 31

Propenyl-methyl ketone, 273
,, pyridine, 116

Propylene, 27, 31, 33, 115, 255

„ addition of hydriodic acid to, 255

dichloride, 27, 115
Proteins, 111, 170, 171
Protocatechuic acid, 117

„ aldehyde, 118
Pseudo-ionone, 104, 105
„ narceine, 157
„ orcinol, 186, 194
„ pelletierin, 29
Pseudo-pelletierin, 29

phellandrene, 69
tropine, 127, 128

,, its conversion into tropine, 128, 129
„ synthesis, 127, 128
uric acid, 162
Pulegone, 67, 68, 94, 95, 96

constitution of, 67 ff.
,, decomposition of, 68
„ formation of, from citronellal, 94, 95
Purine alkaloids, 111, 161 ff., 179
,, derivation of name, 161
„ group, 14

,, „ Fischer's work on, 14

,, ,, nomenclature, 168, 169

„ synthesis, 167

Pyrazole and pyrazoline, derivatives in polymethylene syntheses, 25
Pyridine, 14, 111, 115, 116, 117, 121, 123, 180, 182, 189, 193
Pyridine alkaloids, 115 ff.
„ constitution, 14
,, Ladenburg's synthesis of, 115
,, occurrence of, in alkaloids, 111
,, Bamsay's synthesis of, 115
Pyridinium methyl iodide, 116
Pyridylpyrrol, 122, 123

Pyrones, 180, 182, 191, 192, 193, 194, 125, 196, 197, 200
Pyrrol, 111
Pyrrolidine alkaloids, 112, 120 ff.

316 INDEX OF SUBJECTS

Pyruvic acid, asymmetric synthesis of lactic acid from, 245
„ ester, 265, 266

QUATERNARY ammonium salts, 272
Quinine, 139, 140

„ constitution of, 139 ff.
Quinoles, 201 ff., 235

acetylation of, 217
constitution of, 213 ff .
definition of, 201
preparation of, 205 ff.
properties of, 209 ff.
„ rearrangements of, 215 ff.
Quinoline, 125

„ alkaloids, 134
" Quinoline half " of cinchonine, 134
Quinone, 201, 207, 215, 282

,, isomeric forms of, 266
„ monoxime, 288
Quinonoid and benzenoid character, 215

,, nucleus, effect of substitution on, 215

RACEMATES, resolution of, 246 ff.
Racemization, resolution by, 250
Reactive and non-reactive bodies, 179
Reactivity of ketones, 263

„ variation of, 287

„ and constitution, 261 ff.

„ and physical properties, 285, 287
Rearrangement of affinity, 265
Refractive index, 15, 32, 188, 193, 271, 285
Resolution by amide formation, 250

„ ,, differential racemization, 250
„ „ esterification, 249
„ „ hydrolysis, 250

Resolution methods, modifications of Pasteur's, 247
Rhodinal, 96, 100, 101, 102

„ production of menthone from, 101
Rhodinic acid, 96, 100, 101
Rhodinol, 96, 100, 101
Ring formation, effects of, 31 ff.
Rosaniline, 226
Rule, Markownikofi, 59

SABATIEB and Senderens' reaction, 22, 28, 30, 34

,, „ „ „ applied to polymethylene syntheses, 22

Sandmeyer's reaction, 226
Sebacic acid, 30

" Second half" of cinchonine, 134, 135, 137, 138
Selenium compounds, optically active, 13
Semicarbazide, 212, 213, 274
Sesquiterpene from isoprene, 93
Sesquiterpenes, 39
Silent electric discharge, 38
Silicon compounds, optically active, 13
Silver oxide, 114
Sobrerol, 85, 86, 87
Sobrerythrite, 85, 86
Soline, 111

Solvent, action of optically active, 247, 260
Sorbose bacterium, 247
Spacial arrangement of atoms, 10, 11, 12 ff.

INDEX OF SUBJECTS 317

Spectra, absorption, 15, 265

Stability of polymethylenes, 33, 34

Static and dynamic views of addition reactions, 263

organic chemistry, 10, 16
Stereochemistry, development of, 12 ff.
Stereoisomerism, 12 ff., 272
Steric hindrance, 13
Stewart's theory, 263 ff.
Stilbene, 262
Strain theory, 36, 288
Strecker's method, 171
Styrolene derivatives, 263
Suberane, 29
Suberic acid, 29, 30, 124
Suberone, 29, 124
Suberyl alcohol, 29, 124

iodide, 29, 124, 125

Substitution, effect of, on addition reactions, 261 ff.
Sugars, relations of, to keten group, 197
„ Fischer's work on, 13

production of, from formaldehyde, 178
Sulphothiocarbonic acid, 90

Sulphur, attempted asymmetric synthesis of quadrivalent, 246
compounds, optically active, 13

dioxide, conductivity of triphenyl methyl salts in, 225
Sulphuric acid, heat of addition to trimethylene, 33

„ dehydration with, 50
Sylvestrene, 60
Synthetic chemistry, 15

TAETAEIC acids, asymmetric synthesis of, 246
„ light absorbed by, 251

Pasteur's researches on, 12, 246
Tautomeric bodies, reactivity of, 263
Tautomerism, 9 ff.
Terebic acid, 43, 44

„ synthesis of, 44
Terpenes, 11, 38 ff., 179

Baeyer and Wallach's researches on, 11
classification of, 38
dicyclic, 72 ff.

„ general properties of, 39
,, monocyclic, 38 ff.
,, nomenclature of, 38

olefinic, 91 ff.
Terpenogens, 91
Terpenylic acid, 43, 85

„ „ synthesis of, 44

Terpin, 52 ff.

„ dibromide, 53, 54
„ hydrate, 53
Terpinene, 50 ff., 55

Terpineol, 39 ff., 44, 49, 55, 57, 88, 89, 106, 108
„ decomposition of, 41 ff.
„ dibromide, 85
„ synthesis of, 39 ff.
Terpinolene, 50, 55
Tetracarbonimide, 165
Tetracetic acid, 190, 191, 197
Tetrabromo-cresol-tJ/-bromide, 208

,, ethyl-quinole, 216
Tetrachlorocresol, 204

3i8 INDEX OF SUBJECTS

Tetrachloroquinone, 202
Tetradecapeptide, 176
Tetrahydrotoluic acid, 40, 57
Tetramethylene, 27, 28, 31, 33, 34, 35, 36
„ ' preparation of, 27

carboxylic acid, 27, 28-
,, amine, 28

,, -trimethyl-ammonium hydroxide, 28

Tetrapeptide derivative, 173
Tetraphenyl-ethane, 230

ethylene, 262
Theobromine, 166, 167

„ synthesis, 166

TheophyUine, 165, 166

„ synthesis, 165

Thermochemistry of the polymethylenes, 32 fi.
Thiele's benzene formula, 260
„ theory, 51, 258, 268
„ „ exceptions to, 260, 261

Thioacetic anhydride, 187
Thionyl chloride, 173, 174
Thujenes, 90
Thujone, 90

„ constitution of, 90
Thujyl alcohol, 90
Thymol compound, 198
Tiglic acid, 261

Tin compounds, optically active, 13
Toluidine, 203
Tolyl-hydroxylamine, 205
Transition temperatures, 247
Trans-terpin, 53, 54
Triacetic acid, 190, 191, 197

„ lactone, 186, 190, 191
Tribromacrylic acid, 262
Tribromotriphenyl carbinol, 226
-methyl, 227

chloride, 226, 233
Tribromoxylo-quinole, 201
Trichloracetic acid, 115
Trichlorohydrin, 115
Trichloropurine, 167
Trigonelline, 119, 120
Trihydroxyhexahydrocymene, 42, 43, 47
Trimethylbromocyclopentane carboxylic ester, 74
Trimethylacrylic acid, 261
Trimethylamine, 30, 92, 125, 147, 270 •
Trimethylene, 21, 27, 31, 33, 35, 36
,, bromide, 115

„ chloride, 27

,, preparation of, 27

Trimethylsuccinic acid, 81
Tripeptide, 175

,, derivative, 173
Triphenyl-bromo-methane, 223
carbinol, 229

chloro-methane, 224, 230, 232
methane, 230
methyl, 222 ft.

„ bromide, 225

chloride, 224, 225
„ double compounds, 224

INDEX OF SUBJECTS 319

Triphenyl-methyl, hexaphenyl-ethane view of, 228 ff.
iodide, 224

molecular weight of, 226, 229
peroxide, 224, 229, 235
preparation of, 223
properties of, 223
quinonoid views of, 231 fi.
salts, conductivity of, 225
tautomerism, view of, 237 ff.
trivalent carbon, hypothesis of, 225 ff .
two forms of, 238
Tropic acid synthesis, 130
Tropidine, 124, 126, 127, 129
„ methyl bromide, 126
,, synthesis, 124 ff.
Tropine, 127, 129, 131

„ and pseudotropine, isomerism of, 128
„ synthesis, 127 ff.
Tropinone, 29, 127, 128, 129, 130, 132
„ carboxylic acid, 132

derivatives, 130, 132, 133
„ sodium salt, 132
" True " terpenes, 38
Tube of force, 288
Type theory, 2, 3

ULLMANN and Borsum's hydrocarbon, 228, 239, 230, 235

Umbelliferone compound, 198

Unsaturated body, definition of, 269

Unsaturated ketones, reactions with hydroxylamine, 67, 275 ff .

Unsaturation, 269 ff.

and isomerism, 12

degrees of, 272

its chief effects, 280

variability of, 279

Uramil

162

Urea, 162, 163, 164

Uric aci£, 161, 162, 163, 164, 165, 167
,, constitution, 162
„ decompositions, 164, 165

salts, 163, 167
,, syntheses, 161 ff.

Uroxanic acid, 165

VALENCY, variability of, 278 fi.

Valerianic acid, asymmetric synthesis of, 244

Vanillin, 143

„ methyl ether, 146
Vestrylamine hydrochloride, 60
Vibration hypothesis of benzene, 6 ff., 10

,, intramolecular, 8
Vinyl-acetic acid, 284
Violet perfume, artificial, 104
Violuric acid, 162
Vorlander Rule, 277 ff.

„ theory, 255 ff. -

WANDERING of groups in quinoles, 215 ff.

320 INDEX OF SUBJECTS

XANTHINE, 168
Xanthogenic acid, 90
Xylite, 248
Xyloquinole, 212, 219

ZEISEL'S method, 113, 114

Zincke's researches on chlorination, 202 fi., 207

THE END

PKINTED BY WILLIAM OLOWES AND SONS, LIMITED, LONDON AND BECCLES.

l<i DAY USE

TO DESK FROM WHICH BORROWED
LOAN DEPT.

This book is due on the last date stamped below, or

on the date to which renewed.
Renewed books are subject to immediate recall.

3CMayJojjj

SENtONILL

SEP 1 3 2004

U. C. BERKELEY

General Library
LD 21 A— 50m-8,'57 University of California
(C8481slO)476B Berkeley

2 1 829

Qt>
2 s

S77

THE UNIVERSITY OF CALIFORNIA LIBRARY

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Texts

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Recent advances in organic chemistry"

See other formats